Episode 55, “A Visit with Stephen Senn: Time, Concurrent Controls, and the Bayesian Guidance”
On the latest episode of "In the Interim...", Dr. Scott Berry sits down with award-winning statistician Dr. Stephen Senn for a rigorous discussion on adaptive trial design, platform trials, and the realities behind statistical modeling.
Key issues discussed:
● Regression to the mean is central to valid inference
● Concurrent control in platform trials is not always sufficient
● Confounding between time and treatment must be modeled directly
● Dichotomizing endpoints is “insane”—leads to major information loss
● DSMB oversight grows more challenging with adaptive designs
● Blinding and randomization are critical protections against bias
This conversation challenges assumptions and digs into what really drives trial efficiency.
Tune into the full episode today: https://www.berryconsultants.com/resource/55-a-visit-with-stephen-senn-time-concurrent-controls-and-the-bayesian-guidance