Skip to main content

Table 2 Effect of ICU admission on in-hospital mortality using standard logistic regression (crude and adjusted logistic models) and instrumental variable-based analyses (double-stage logistic regression and double-stage probit structural equation model)

From: Benefits of ICU admission in critically ill patients: Whether instrumental variable methods or propensity scores should be used

 

Odds Ratio

95CI

p-value

Crude Logistic Regression

0.74

0.65-0.84

< 0.01

Adjusted Logistic Regression

1.25

1.07-1.46

0.01

Propensity matched cohort

1.23

1.04-1.45

0.01

2LR

0.73

0.24-2.45

0.56

Probit Model

0.89

0.24-2.37

0.71

  1. The association measure is the odds ratio (with 95% confidence interval, 95CI). 2LR: double stage logistic regression