Monitoring and evaluating clinical trials data.

Abstract

This paper presents an overview of recent developments in statistical methods for the design of clinical trials. Whilst the methods are suitable for application in a wide range of therapeutic areas, they are of particular value in intensive care medicine due to its high levels of mortality and short periods of follow-up. The paper reviews the calculation of the sample size necessary to achieve a given power requirement, and then describes the operation of a Data and Safety Monitoring Board, the mid-trial reassessment of sample size and the use of stopping rules to terminate a study as soon as sufficient evidence is available. The purpose of the paper is to encourage investigators to consider the wide range of design options which have recently become available, when planning a new clinical study.