I think this is accurate. A significant loading on diagnosis (X2) does not tell you anything about the effect of diagnosis at any particular age (except age 0)

I think this is accurate. A significant loading on diagnosis (X2) does not tell you anything about the effect of diagnosis at any particular age (except age 0). You’d have to recenter the model about the age of interest.