Skip to main content
. 2019 Jul 16;27:67. doi: 10.1186/s13049-019-0642-2

Table 3.

Selection of variables for inclusion in multivariable models

Predictor variable Incidence of variable selection Wald test p-value
GCS (sum) model GCS (components) model AVPU model GCS (sum) model GCS (components) model AVPU model
Location 50 50 50 0.002 0.004 0.003
Age 50 50 50 < 0.001 < 0.001 < 0.001
Gender 0 0 0
Resps 50 50 50 < 0.001 < 0.001 < 0.001
SpO2 50 50 50 < 0.001 < 0.001 < 0.001
Pulse 50 50 50 < 0.001 < 0.001 < 0.001
SBP 50 50 50 < 0.001 < 0.001 < 0.001
DBP 0 0 0
Temp 50 50 50 < 0.001 < 0.001 < 0.01
BM 1 1 0 0.86* 0.85* 0.87*
Skin 50 50 50 0.007 0.007 0.014
CBRT 0 0 0
RPupilReact 11 12 8 0.42* 0.41* 0.46*
RPupilSize 0 0 0
GCS_sum 50 NA NA < 0.001 NA NA
GCS_eye NA 0 NA NA NA
GCS_verbal NA 50 NA NA < 0.001 NA
GCS_motor NA 0 NA NA NA
AVPU NA NA 50 NA NA < 0.001

SpO2 peripheral oxygen saturation, SBP systolic blood pressure, DBP diastolic blood pressure, Temp temperature (C), BM blood sugar mmol/L, CBRT capillary bed refill time, GCS Glasgow coma score, AVPU alert, verbal, pain or unresponsive

*Did not reach statistical significance