Table 4.

Factors associated with ventilator-associated pneumonia by multivariate analysis

Variables	P	OR (95 % CI)
Model 1
SAPS II	0.019	1.01 (1.00, 1.02)*
Red blood cell transfusion	0.009	1.75 (1.14, 2.70)
Proton pump inhibitor use	0.040	1.86 (1.03, 3.39)
Number of days with hyperoxemia	0.001	1.10 (1.04, 1.16)**
Model 2
SAPS II	0.017	1.01 (1.00, 1.02)*
Red blood cell transfusion	0.009	1.79 (1.16, 2.78)
Proton pump inhibitor use	0.031	1.92 (1.06, 3.50)
MV duration prior to VAP	0.012	1.10 (1.01, 1.04)**
Hyperoxemia at ICU admission	0.004	1.89 (1.23, 2.89)
Model 3
Red blood cell transfusion	0.018	1.71 (1.10, 2.65)
Proton pump inhibitor use	0.017	2.10 (1.14, 3.79)
MV duration prior to VAP	0.005	1.03 (1.01, 1.04)*
Percentage of days with hyperoxemia	0.029	2.20 (1.08, 4.48)**

SAPS simplified acute physiology score; MV mechanical ventilation, VAP ventilator-associated pneumonia. Model 1: *per point of SAPS II, **per day with hyperoxemia; Hosmer-Lemshow goodness-of-fit, p = 0.62. Other non-significant variables included in the model: transfer from other wards, shock, sedation, and duration of MV prior to VAP. Model 2: *per point of SAPS II, **per day of mechanical ventilation; Hosmer-Lemshow goodness-of-fit, p = 0.81. Other non-significant variables included in the model: transfer from other wards, shock, and sedation. Model 3: *per day of mechanical ventilation, **per centile of days with hyperoxemia; Hosmer-Lemshow goodness-of-fit, p = 0.78. Other non-significant variables included in the model: SAPS II, transfer from other wards, shock, and sedation