Skip to main content

Table 2 Effect decomposition for four TDF constructs as hypothesised mediators

From: Mechanisms of implementing public health interventions: a pooled causal mediation analysis of randomised trials

Mediator Intervention-mediator effect Mediator-outcome effect ATE ADE ACME Proportion mediated (%)
Knowledge − 0.84 (−5.05 to 3.37) 1.01 (0.96 to 1.06) 0.40 (0.19 to 0.57)* 0.40 (0.20 to 0.58)* 0.00 (− 0.01 to 0.04) 0.00 (− 0.18 to 0.09)
Skills 2.66 (−1.91 to 7.24) 0.99 (0.96 to 1.02) 0.40 (0.19 to 0.57)* 0.38 (0.18 to 0.56)* 0.02 (− 0.02 to 0.08) 0.05 (− 0.08 to 0.21)
Professional role and identity − 0.88 (− 5.12 to 3.37) 1.00 (0.95 to 1.05) 0.38 (0.17 to 0.55)* 0.38 (0.18 to 0.56)* 0.00 (− 0.03 to 0.03) 0.00 (− 0.09 to 0.09)
Environmental context and resources − 2.12 (− 6.54 to 2.31) 0.99 (0.96 to 1.03) 0.39 (0.19 to 0.56)* 0.38 (0.18 to 0.56)* 0.00 (− 0.03 to 0.04) 0.00 (− 0.10 to 0.11)
  1. All effects unstandardized with their 95% confidence intervals. The mediator-outcome effects are presented as odds ratios
  2. ATE average total effect, ADE average direct effect, ACME average causal mediation effect
  3. *p = < 0.05