Skip to main content
. Author manuscript; available in PMC: 2009 Nov 11.
Published in final edited form as: J Perinatol. 2008 Mar 13;28(6):420–426. doi: 10.1038/jp.2008.18

Table 2.

Variables identified by stepwise logistic regression models as associated with death/BPD. Variables were allowed to enter/stay in the model at p<0.2

Full data set
Stepwise logistic
regression
Variables added Odds ratio ( 95% CI )
First Step Birth weight (per 100g) 0.78 (0.71 – 0.85)
Second Step Mean FiO2 3.14 (0.72 – 13.63)
Third Step Male gender 2.54 (150 – 4.28)
Fourth Step Surfactant dose 1.44 (1.06 – 1.96)
Fifth Step Mean OI (per unit) 1.03 (1.00 – 1.05)
Sixth Step Outborn 1.73 (0.91 – 3.30)
c statistic for model = 0.75
Treatment (iNO) group
Stepwise logistic
regression
Variables added Odds ratio (95% CI)
First Step Birth weight (per 100g) 0.65 (0.56 – 0.76)
Second Step Mean FiO2 26.54 (3.88 – 181.48)
Third Step Age at randomization (hours) 0.97 (0.96 – 0.99)
Fourth Step Surfactant dose 2.21 (1.27 – 3.84)
Fifth Step Seizures 0.28 (0.04 – 1.85)
c statistic for model = 0.84
Control group
Stepwise logistic
regression
Variables added Odds ratio (95% CI)
First Step Male gender 4.67 (2.13 – 10.27)
Second Step Gestational age (per week) 0.86 (0.73 – 1.02)
Third Step High frequency ventilation 0.39 (0.18 – 0.86)
Fourth Step Outborn 2.34 (0.91 – 6.04)
Fifth Step Air Leak 0.33 (0.11 – 0.96)
Sixth Step Mean OI (per unit) 1.02 (0.99 – 1.05)
Seventh Step Change in PaO2 (baseline to 5
ppm study gas)
0.98 (0.96 – 1.00)
Eighth Step PaO2 Max Diff 1.02 (1.00 – 1.04)
c statistic for model = 0.76