Skip to main content

Table 1 Impact of steroid depending on the model and the variable used

From: Replicating finding, answering questions: closer to the truth about COVID-19 associated VAP

 

Derived from the study

Minimal model

Full model

Propensity matching

1.81 [1.31;2.50]

1.27 [1.07;1.51]

1.09 [0.87;1.36]

Competing risk regression

1.28 [1.03–1.58]

1.24 [1.03;1.48]

1.04 [0.87;1.25]

cs Cox model

–*

1.13 [0.93;1.36]

1.12 [0.93;1.36]

  1. Cs: Cause specific, Minimal model: including the variable available used in the original study, Full model: using all variable available
  2. *No csHR estimated for the overall effect of steroid