Data-Analytic Models

Data-analytic models refer to the models that will be used to analyze data from a prospective trial based on the proposed statistical analysis plan (SAP). These models may include ANOVA- or ANCOVA-based models, linear mixed-model repeated measures (MMRM) or nonlinear mixed effects models, to name a few, and typically depend upon the nature of the endpoint. These data-analytic models may or may not be similar to the data generation models used to simulate clinical trial data depending on the objectives of the trial. Longitudinal dose- and/or exposure-response models should be considered for characterizing the shape of the exposure-response relationship whereas ANCOVA-based analyses are typically considered for confirmatory trials in a hypothesis testing framework requiring minimal assumptions such as a Phase 2 registration trial. Note that different data-analytic models can be proposed for the same trial based on the different study objectives.