Abstract:
|
Pragmatic trials are traditionally designed to test the effectiveness of some approach or intervention in real-world practice. As such, they are generally designed to place great emphasis on external validity. While design decisions are made so that the environment of the trial mimics the real world, they also introduce challenges for participant identification, enrollment, randomization, data collection and quality, and systematic collection of outcomes and safety information. These challenges can threaten internal validity, but they do not need to. This talk will discuss the domains of pragmatism and the study design considerations that can help to balance external validity with internal validity, as well as how the statistical analysis plan can mitigate some of the design challenges.
|