What is it about?
The inclusion criteria for a randomised clinical; trial cannot be used as a means of generalising the results because the patients recruited are rarely a representative sample of patients who could be recruited. Using the results from clinical trials requires recognising that they are experiments. Judgements as to how the results should be used rely on theory and the use of appropriate scales and are inevitably potentially fallible.
Featured Image
Photo by milos tomasevic on Unsplash
Why is it important?
The currently fashionable topic of estimands has been useful in clarifying how to deal with intercurrent events. However, the notion of traget populations is in danger of leading researchers astray by encouraging the false belief that some sort of representative sampling takes place. It does not and cannot.
Perspectives
I know many of the authors and they are experts in this field. If anybody could make success of applying the notion of randomization theory to target populations they could. The fact that they don't succeed is a warning that something is wrong with the basic idea.
Professor Stephen J Senn
Consultant Statistician
Read the Original
This page is a summary of: Comment on “Using Randomization Tests to Address Disruptions in Clinical Trials: A Report from the NISS Ingram Olkin Forum Series on Unplanned Clinical Trial Disruptions”, Statistics in Biopharmaceutical Research, October 2024, Taylor & Francis,
DOI: 10.1080/19466315.2024.2375232.
You can read the full text:
Resources
Contributors
The following have contributed to this page
