A group sequential design is a type of adaptive design where the number of patients isn’t set in advance. Patients are divided into an equal number of groups and data is analyzed at pre-determined points in the trial. A stopping rule specifies when and why a trial might be halted at these points.
- As soon as positive results are found in the trial (for example, a drug shows statistically significant benefits), the trial is continued.
- As soon as negative results are found (for example, adverse effects), the trial is ended, so the trial could be stopped at any point in time.
Interim Analysis in Group Sequential Design
The circumstances under which the trial will be stopped or continued is specified in advance (Kelly et. al, 2005). A group sequential design includes a pre-determined number of stages, including interim stages (with the associated interim analysis) and a final stage. Each stage is specified by:
- The sample size,
- Critical Values,
- A stopping criterion to either support or reject the null hypothesis.
At the end of each interim stage, data analysis is usually performed or reviewed by a Data and Safety Monitoring Committee (DSMC). Interim analysis involves calculation of a test statistic. This is compared to the critical value to decide whether to stop or continue the trial.
Comparison to Fixed Sample Designs
The typical fixed sample design has patients entering the trial sequentially in matched pairs for two different treatments; Data is only analyzed once at the conclusion of the trial. Group sequential has patients entering in groups; Data is analyzed at a certain number of specified stopping points.
Kelly et. al (2005). A practical comparison of group-sequential and adaptive designs. J Biopharm Stat. 2005;15(4):719-38.
Pocock (1977). Group sequential methods in the design and analysis of clinical trials. Biometrika. 64, 2, pp. 191-9.
Chow & Lu. (2008). Design and Analysis of Clinical Trials: Concepts and Methodologies.