MMRM vs. LOCF: a comprehensive comparison based on simulation study and 25 NDA datasets.
Missing data are ubiquitous in longitudinal clinical trials, and the impact on power has been extensively assessed in the literature. Multiple doses of the investigational product and multiple efficacy endpoints are often studied in randomized clinical trials and multiplicity adjustment needs to be considered in the sample size calculations. In this paper, I show how to perform sample size calculations with multiplicity adjustment for longitudinal clinical trials with missing data by converting longitudinal data with missing data to cross-sectional data without missing data. The proposed approach can drastically simplify the simulation work and facilitate the evaluation of power for various scenarios.