How to Define "astar," "a," and "mval" for Continuous Exposure and Mediation?

[lywbetter]

Hi,
My mediation analysis setup includes continuous exposure, continuous mediation, and a time-to-event outcome. I have some questions about defining the values for "astar," "a," and "mval":

astar: The control value for the exposure (default is 0).
a: The active value for the exposure (default is 1).
mval: A list specifying a value for each mediator variable at which the variable is controlled (used when the model is rb, wb, ne, msm, or formula).
I've tested my data with mval set to list("0"), list("1"), and list("100") separately, but the results are the same in all cases.

Can I safely ignore this parameter in my analysis?

Thank you for your guidance!

[Yijlor]

I also have the same question. When I tested mval for list(0) and list(1), the point estimate for the "Direct counterfactual imputation estimation" remains the same, but it seems the 95% CI changes.