6. Consider a factorial randomized trial of two interventions (A and B) and an outcome D, in which a third variable Z is observed and measured at baseline and does not change over time. Z is not an effect of D. Assume perfect compliance and adequate statistical power. Without further assumptions, can I assess whether Z is an effect measure modifier of the causal effect of B on D? Why or why not? (Be brief! And recall, effect measure modification is what we assess in a typical subgroup analysis.)