Refined Moderation Analysis with Binary Outcomes
With the growing interest in personalized or precision medicine, it is indispensable that moderation analysis which is primarily related to the study of differential treatment effects among patients with different characteristics, also serves as the bedrock for precision medicine is taken more seriously. Concerning moderation analysis with binary outcomes, we start with an interesting observation, which shows that heterogeneous treatment effects could be equivalently estimated via a role exchange between the outcome and the treatment variable. The result holds for both experimental data and observational data, yet with an important difference in interpretation. Two estimators of moderating effects corresponding to two GLM models can be obtained. We combine the two models into a single model and employ the GEE approach to simultaneously obtain parameter estimates associated with the moderating effects (interaction terms), on which basis of refined inference can be made. The improved efficiency is helpful in addressing the lack-of-power problem that is common in the search for moderators. We investigate the proposed method by simulation and provide an illustration with data from a randomized trial concerning wart treatment. Essentially, the new revelation about the ‘role swapping’ technique can be useful by offering more flexible and computational ease in scenarios where direct modeling of interactions encounters difficulties and becomes inconvenient, owing to modeling complexity, numerical difficulty, or unavailability of implementation.
Anto, Eric, "Refined Moderation Analysis with Binary Outcomes" (2021). ETD Collection for University of Texas, El Paso. AAI28542981.