Interim analysis in long-term clinical trials

The purpose of this dissertation is to evaluate the usefulness of both stopping rules and estimation methods in long-term clinical trials with interim analyses. The ASPECT trial, a long-term clinical trial to assess the effect of anticoagulant therapy on mortality in patients after myocardial infarction which is currently conducted in the Netherlands, serves as a major example throughout this dissertation. In the final stage of the development of a new therapy usually a large-scale comparative clinical trial is used. In many clinical trials the effect of a treatment can only be assessed over a long period of time, or the length of time that is needed to recruit the required number of patients for the trial is long. During the course of such long-term clinical trials it is desirable to evaluate the accumulating results at regular time intervals. The interim results might indicate that continuation of the trial is meaningful, superfluous, or ethically unjustified. This dissertation describes statistical methods for the design and the analysis of long-term clinical trials with interim analyses within the classical statistical framework. The usefulness of these methods is investigated. The ASPECT trial, a trial that is currently being conducted in the Netherlands of the effect of anticoagulant therapy on mortality in patients after myocardial infarction, serves as a major example throughout this dissertation.