HOW RELIABLE ARE CLINICAL-TRIALS - THE IMPORTANCE OF THE CRITERIA FOR EARLY TERMINATION

The early termination of clinical trials, for either benefit or harm, often generates undue enthusiasm or alarm. The enhanced publicity attending early termination of a trial promotes inappropriate interpretations that are aided by the inherent difficulty of prompt and comprehensive data review. Furthermore, the process of monitoring the accumulating outcome data for early evidence of treatment benefit or harm is fraught with many statistical methodological difficulties. This article is based on a report from a Task Force convened by the Working Group on Arrhythmias of the European Society of Cardiology. It reviews, as examples to illustrate the importance of suitable trial design and of proper stopping rules, a series of trials terminated appropriately or inappropriately for benefit or harm; it analyses the general design issues and it reviews the main issues in interim analysis with special reference to main strategies for reducing the rate of false-positive claims that could result from early trial termination.