In the leukemia trial [ 9], multiple amendments, most notably amending the primary endpoint, were made subsequent to interim analysis with knowledge of treatment-arm specific rates in an apparent attempt to increase the number of events accrued and achieve a successful trial, but a substantial bias was introduced.
However, a recommendation for SSR made with knowledge of interim data introduces bias to the trial.
Here 'promising' means that the conditional power at the current parameter estimate, i.e. the power updated by the accumulated knowledge at interim, amounts to at least 50%.
Obviously, reporters sat on this knowledge in the interim, never investigating the claim further because they are corrupt toadies to power the matter had no real-world relevance.
Bayesian adaptive clinical trial design was developed from sequential designs in which the design can be changed based on knowledge gained from interim analyses.
Despite receiving information on how to stay healthy, the patient did not change his behaviour, or even increase his knowledge as an interim step.
In general, to minimize the potential for bias, the trial leadership, which is insulated from knowledge of the interim data, rather than the DMC, should be responsible for proposing potential changes other than those driven by safety considerations".
Changing the primary endpoint of a study in the knowledge of the interim results is analogous to changing a bet to the winning horse part way through a race, introduces bias and leads to complications with interpretation of the final results.
We surmise that this strategy is justified ad interim given the current state of knowledge in systems biology in vivo.
Note, however, that the posterior probabilities are only updated at each interim analysis, as they require full knowledge of patient outcomes.
