Improving the Information Content of RCT Endpoints
View Presentation View Presentation
*Vance Berger, NIH 

Keywords: dimensions, composite, information-preserving,

There are several dimensions of most diseases. A given experimental treatment may improve any or all of these, relative to the control treatment. Often, even at the time of initiation of the pivotal Phase III study, it will not be clear which of these dimensions (endpoints) will provide the best opportunity to show the treatment effect. Even if it is clear that the novel treatment will not improve certain dimensions, these are still part of the equation, because they should not worsen. It is common to analyze each of a wide variety of endpoints separately, and adjust or pick one as primary. In part because adjustments may be conservative, it is quite common to pick a primary endpoint which may be a composite endpoint. Unfortunately, when referring to a composite endpoint for combining binary endpoints, with no further elaboration, the understanding is generally that all component b