Skip to main content

Table 2 Estimates of total, direct and indirect effects of maternal depression or anxiety in pregnancy on the risk of infant wheezing between 6 and 18 months of age from inverse odds ratio weighting, inverse probability weighting and imputation approach. \(M_{1}\): adverse reproductive outcomes. \(M_{2}\): infant lower respiratory infections

From: Applied causal inference methods for sequential mediators

 

Through \(M_{1}\)

Through \(M_{1}\) and \(M_{2}\)

 

PR

95% CI*

PR

95% CI

Conditional effect

IOR\(^{*}\) approach

     Direct effect

1.59

1.27-1.94

1.57

1.25-1.92

     Indirect effect

1.03

0.94-1.12

1.05

0.95-1.15

     Total effect

1.64

1.33-2.00

1.64

1.33-1.97

Marginal effect

IPW\(^{**}\) approach

     Direct effect

1.60

1.30-1.94

1.57

1.27-1.87

     Indirect effect

1.02

0.99-1.04

1.04

0.99-1.09

     Total effect

1.63

1.33-1.98

1.63

1.31-1.95

Conditional effect

Imputation approach

     Direct effect

1.60

1.31-1.94

1.57

1.26-1.90

     Indirect effect

1.02

1.01-1.05

1.05

1.01-1.09

     Total effect

1.64

1.33-1.99

1.64

1.33-1.99

Marginal effect

Imputation approach

     Direct effect

1.60

1.30-1.91

1.57

1.24-1.88

     Indirect effect

1.02

1.00-1.04

1.04

0.99-1.09

     Total effect

1.63

1.33-1.95

1.62

1.29-1.95

  1. PR: prevalence ratio; CI: confidence interval calculated by bootstrap.
  2. * Inverse odds ratio weighting.
  3. ** Inverse probability weighting