Skip to main content

Table 4 Subgroup analyses describing the association of the protocol periods with severity-adjusted hospital mortality

From: Impact of introducing multiple evidence-based clinical practice protocols in a medical intensive care unit: a retrospective cohort study

Subgroup

Variable

Odds ratio (95% CI)

P-value

High severity of illness

   
 

Predicted mortality, %

1.040* (1.036 – 1.043)

< 0.001

 

Protocol period

  
 

   Pre-protocol

Reference

 
 

   Transition

0.831 (0.701–0.984)

0.031

 

   Protocol

0.754 (0.625–0.910)

0.003

Low severity of illness

   
 

Predicted mortality, %

1.241* (1.191 – 1.293)

< 0.001

 

Protocol period

  
 

   Pre-protocol

Reference

 
 

   Transition

1.013 (0.730–1.406)

0.938

 

   Protocol

0.733 (0.492–1.093

0.128

Long ICU length of stay

   
 

Predicted mortality, %

1.033* (1.029 – 1.036)

< 0.001

 

Protocol period

  
 

   Pre-protocol

Reference

 
 

   Transition

0.893 (0.746–1.069)

0.217

 

   Protocol

0.745 (0.609–0.910)

0.004

Short ICU length of stay

   
 

Predicted mortality, %

1.077* (1.071 – 1.083)

< 0.001

 

Protocol period

  
 

   Pre-protocol

Reference

 
 

   Transition

0.829 (0.620–1.109)

0.207

 

   Protocol

0.953 (0.686–1.325)

0.776

  1. CI = Confidence interval
  2. *This odds ratio reflects the risk of death per 1% rise in the predicted mortality rate for the entire study population.