Skip to main content

Table 3 Multivariable analysis of factors associated with 30-day mortality

From: Development of a new score for early mortality prediction in trauma ICU patients: RETRASCORE

Variable

β-coefficientsa

OR (95% CI)

p value

Points

Age groups

    

 < 50

 

Reference

 

0

 50–65

0.598

1.83 (1.34–2.47)

< 0.001

1

 65–75

1.239

4.78 (3.43–6.66)

< 0.001

2

 > 75

2.198

10.35 (7.47–14.36)

< 0.001

4

PRECOAG

0.349

1.43 (1.10–1.85)

0.007

1

Pre-hospital intubation

0.336

1.41 (1.08–1.83)

0.012

1

High Risk mechanism

0.662

1.96 (1.56–2.47)

< 0.001

1

PUPILS

    

 Normal

 

Reference

 

0

 Unilateral mydriasis

0.950

2.63 (1.97–3.51)

< 0.001

2

 Bilateral mydriasis

3.217

26.38 (17.51–39.74)

< 0.001

6

GCS ≤ 8

0.841

2.35 (1.79–3.10)

< 0.001

2

MAIS-Head

0.495

1.66 (1.30–2.12)

< 0.001

1

MAIS-Thorax

− 0.271

0.76 (0.60–0.96)

0.026

− 1

Haemodynamic failure

1.148

3.21 (2.51–4.12)

< 0.001

2

Respiratory failure

0.708

2.06 (1.58–2.67)

< 0.001

1

Coagulopathy

0.567

1.78 (1.36–2.33)

< 0.001

1

Mechanical ventilation

0.580

1.80 (1.39–2.34)

< 0.001

1

Massive haemorrhage

0.452

1.58 (1.10–2.27)

< 0.001

1

  1. Logistic regression. Derivation set (n = 5976)
  2. PRECOAG, prior treatment with antiplatelets or anticoagulants; GCS, Glasgow Coma Score; AIS, Abbreviated Injury Scale; MAIS, AIS ≥ 3
  3. aRegression coefficient multiplied with a shrinkage factor (bootstrapping procedure) of 0.98