Eligibility exclusions, losses to follow-up, removal of randomized patients, and uncounted events in cancer clinical trials

Abstract

This paper reviews the potential impact of eligibility exclusions, losses to follow-up, removal of randomized patients from the analyzed data, and failure to count all observed events on the interpretation of cancer clinical trials. The pragmatic and explanatory approaches to these problems are presented, and it is argued that most clinical trials yield data that are suitable for answering pragmatic questions. Thus, the pragmatic approach to these problems is stressed.