Skip to main content

Table 4 Multiple linear regression analysis of independent predictors of outcomes in children with sepsis, and with septic shock

From: Timing of antibiotics, volume, and vasoactive infusions in children with sepsis admitted to intensive care

Variable

PICU days

Ventilator days

 

Effect size

95 % CI

P value

Effect size

95 % CI

P value

PRISM

0.64

0.23, 1.04

0.003

0.30

0.01, 0.59

0.04

Age

0.015

-

0.89

−0.02

-

0.22

Severe underlying disease

7.27

1.34, 13.2

0.017

4.1

-

0.05

Time to antibiotics, minutes

−0.010

-

0.92

0.002

-

0.45

Model R2

15.2 %

  

7.4 %

  

Subgroup: septic shock

   

PRISM

0.71

0.16, 1.25

0.012

0.24

0.03, 0.45

0.024

Age

0.05

-

0.30

−0.22

-

0.12

Severe underlying disease

0.29

-

0.06

3.82

0.57, 1.07

0.023

Time to antibiotics, minutes

-.067

-

0.63

0.049

-

0.72

Volume in 2 h

0.22

0.05, 0.38

0.010

0.09

0.02, 0.15

0.009

Volume prior to inotropes

−0.19

-

0.38

−0.13

-

0.54

Model R2

23.4 %

  

32.1 %

  
  1. Analyses were performed using stepwise multiple linear regression, except for ventilator days in the entire cohort, where the variables were forced into the model. For the drop in pediatric logistic organ dysfunction score (PELOD) from day 1 to 3 of sepsis, the only independent predictor in the entire cohort was age (effect size 0.042; 95 % CI 0.004, 0.080; p = 0.032); for the septic shock subgroup, there were no independent predictors. PRISM pediatric risk of mortality score