Skip to main content

Table 3 Multivariate regression models for association of outcome with DFR score after controlling for APACHE II score.

From: A Genomic Score Prognostic of Outcome in Trauma Patients

 

APACHE II score

DFR score

Model significance

Outcomea

   

Days in ICU

2.2±0.9 (0.02)

2.2±0.9 (0.02)

< 0.001

Days on ventilator

2.1±0.7 (0.002)

2.3±0.7 (< 0.001)

< 0.001

Hospital length of stay

2.6± 1.5 (0.08)

3.6± 1.5 (0.02)

0.002

Maximum Marshall score

0.7±0.2 (< 0.001)

0.8±0.2 (< 0.001)

< 0.001

Maximum Denver score

0.5±0.2 (0.002)

0.5±0.2 (< 0.001)

< 0.001

Eventb

   

Mortality

1.9 (0.7,7.0) (0.19)

1.5 (0.7,3.1) (0.32)

0.09

MODS (Marshall criterion)

1.7 (1.1,2.4) (0.009)

1.4 (0.9, 1.9) (0.09)

< 0.001

MODS (Denver criterion)

1.4 (0.8,2.3) (0.24)

1.9 (1.2,3.0) (0.01)

0.001

Nosocomial infections

1.9 (1.3,2.7) (0.001)

1.4 (0.95,2.0) (0.09)

< 0.001

Surgical site infections

1.7 (1.0,2.6) (0.03)

1.2 (0.8, 1.8) (0.42)

0.02

  1. aThe model is Outcome = Intercept + α APACHE II + β DFR + ε. Table entries show the estimated regression coefficients, their standard errors, and significance. The means and standard deviations used to standardize scores were given in Table 2.
  2. bFor event outcomes, the model is Log Odds (Event)) = Intercept + α APACHE II + β DFR. Each table entry shows the estimated odds ratio, its 95% Cl and its significance. The odds ratio indicates the increased (or decreased) odds of the event for one unit increase in the standardized score when the other variable is held fixed.
Back to article page

Molecular Medicine

ISSN: 1528-3658