Dear Sara,

There are abundant different (longitudinal) mediation methods to choose from.

The first thing to consider is that you ensure the satisfaction of the non-parametric assumptions (e.g., uncounfoundedness assumptions), regardless of your parametric model.

Although you have an RCT in your data, you can still have mediator-outcome confounders (time-varying and time-constant) that may strongly impact your mediation analysis. So be aware of these and adjust them accordingly.

Second, regarding longitudinal mediation, I need to ask which measures you mean with:

"I have one measure before starting acceptance commitment therapy (in an RCT), 6 months after, 12 months after, and now years after."Do you have multiple measures for Pain Catastrophizing(M), Psychological Flexibility (X), Physical Functioning (Y) at four points in time?

Unfortunately, I am not an expert in latent difference score mediation.

However, I would recommend looking at the article by Selig & Preacher (2009):

Selig J.P. & Preacher K.J. (2009). Mediation models for longitudinal data in developmental research. Research in Human Development, 6(2–3), 166–164. https://doi.org/10.1080/15427600902911247

Personally, I would recommend applying the mediational g-formula of VanderWeele & Tchetgen Tchetgen (2017) if you have time-varying variables for X, M and Y since I am unaware of the capabilities to deal with time-varying confounding by latent difference score mediation. Further, I would assume that violation of the fourth uncounfoundedness assumption (exposure-induced mediator-outcome confounding) could be present.

For a comprehensive introduction, see:

VanderWeele, T. J., & Tchetgen Tchetgen, E. J. (2017). Mediation analysis with time varying exposures and mediators. Journal of the Royal Statistical Society, Series B, 79(3), 917–938. https://doi.org/10.1111/rssb.12194

and for an application in Stata, you might take a look at my publication

Demirer, I., Kühhirt, M., Karbach, U., & Pfaff, H. (2022). Does positive affect mediate the association of multimorbidity on depressive symptoms?. Aging & mental health, 26(1), 65–76. https://doi.org/10.1080/13607863.2020.1870209

Thank you a lot for your in-depth answer. I will look into the sources you listed, as well as your article- congratulations on your publication. Yes, I have measures for all, at four different points in time.

S

Meditation joins with the power of prayer supported by spiritual mantra which may help to keep our mind calm & quiet in midst of worries , tension & frustration . With this some years back I have also expressed my views reading as under

''As it is thru our self we may reach to higher spirit of our life in the form of religion , the power of prayer ,Meditation ,& self breathing to keep us in our silent zone

This is my personal opinion