Skip to main content

Table 1 Overview of the different scenarios. In the scenarios with carry-over, the parameters for wash-in \(\tau\) and wash-out \(\gamma\) simulating the exponential decay are changed. \(w_{A,T}\) denotes an effect modifier of Activity on Treatment. Hence, the treatment effect depends on Activity whenever \(w_{A,T}\ne 0\)

From: Comparison of Bayesian Networks, G-estimation and linear models to estimate causal treatment effects in aggregated N-of-1 trials with carry-over effects

 

No Activity interaction

Activity interaction

 

Scenario 1

Scenario 3

No carry-over

\(\gamma =\tau =1\)

\(\gamma =\tau =1\)

\(w_{A,T} = 0\)

\(w_{A,T} \ne 0\)

 

Scenario 2

Scenario 4

With carry-over

\(\gamma _{1} = 3\); \(\tau _{1} = 6\)

\(\gamma _{1} = 3\); \(\tau _{1} = 6\)

\(\gamma _{2} = 4\); \(\tau _{2}=5\)

\(\gamma _{2} = 4\); \(\tau _{2}=5\)

\(w_{A,T} = 0\)

\(w_{A,T} \ne 0\)