Interim Analyses

Abstract

Clinical trials tend to have a long duration, because mostly patients are enrolled one by one, and their responses to treatment are observed sequentially. For the organizers this part of the trial is an exciting phase because after all the hard work involved in planning and getting the trial started, finally concrete data will become available. Immediately, there is the possibility to look at the data in order to check that the trial protocol is pursued appropriately by the investigators and to look at any difficulties, e.g., those with patient and/or doctor compliance, and to see whether there is any need for protocol alterations¹. “Looking at the data” for such purposes should, however, be done carefully. In this chapter we will discuss questions such as:

• 1. why should we monitor a trial;

• 2. who should monitor a trial;

• 3. what should be monitored;

• 4. why should we be careful.