Group Sequential Design

Group sequential design is a type of adaptive design where trial can be stopped early based on the results of interim analyses concerning the efficacy or futility of the trial.

In this design, number of study subjects are not fixed in advance. Interim analysis is done to decide upon the required changes in the trial. At all interim points, data are collected and analysed to check whether the trial should continue or stop. (An interim analysis is an analysis of data that is conducted before data collection has been completed.)

Stopping rule

Stopping rule is something which states when and why the trial need to be stopped at particular points. For example – positive outcome in the trial (e.g., drug shows statistically significant benefits) may suggest to continue the trial. Whereas, negative outcomes in the trial (e.g., any adverse effects) may suggest to stop the trial. Stopping rule which is generally set prior to or shortly after the trial starts recruitment of study subjects.

A group sequential design explains detailed specifications for a group sequential trial. It provides pre-determined total number of stages i.e., the number of interim stages plus a final stage. Apart from usual specifications it also explains the following points at each stage:

  • The sample size
  • Critical Values
  • A stopping criterion to accept/reject the null hypothesis

In each interim stage, data is analysed and the values of test statistics is compared to the critical value to decide whether to stop or continue the trial. [A test statistic is a statistic (a quantity derived from the sample) used in statistical hypothesis testing.]

Sub-types of Group sequential design

Advantages of group sequential design
  • Furnishes the pre-determined total number of stages i.e., the number of interim stages plus a final stage
  • Helpful in reducing the needed sample size along with maintaining the power of study and controlling the type 1 error
  • Presents stopping criterion to reject/accept the null hypothesis at each interim stage
  • Presents critical values and the sample size at each stage for the trial Helpful in saving money, resources and time (e.g., early efficacy/futility discovered)
  • Provides a conclusion much earlier compared to a traditional design
Disadvantages of group sequential design
  • Possess complex methodology
  • Need efficient logistics and trained staff
  • Required good guidance, proper coordination among staff and rigorous monitoring

Comparative study of Group sequential design & Fixed Sample Design

The typical fixed sample design trial has a fixed sample size in advance. This design uses the same number of subjects regardless the fact that the true treatment effect is significantly favourable, marginal or really unfavourable relative to the control arm.  The main objective of this design is to accept/reject null/alternative hypothesis.

On the other hand, in group sequential design, number of study subjects are not fixed in advance. Study subjects enter the trial into a number of equal-sized groups in sequential manner depending upon the requirement after each interim analysis.  At predetermined points, gathered data is analysed to study the safety and efficacy profile of the investigational product and to estimate the required changes (e.g., change in number of study subjects) in the trial. At all interim points, data analysis is conducted to decide whether to continue or stop the trial.