Drug Evaluation Committee Study on Statistical Inference of Adaptive Design
Data Science Subcommittee
February 2023
Guidance from the FDA on clinical trials applying adaptive design will be finalized in 2019, and discussions are underway at ICH-E20 with the aim of developing uniform guidelines.
Adaptive design is characterized by the ability to modify the design of a clinical trial mid-stream, which is expected to benefit trial participants by increasing their chances of receiving better treatment, more efficient drug development, and better use of available resources. On the other hand, if statistical methods are not used appropriately, undesirable phenomena may occur in terms of statistical validity, such as an increase in the probability of Type 1 errors, the occurrence of bias in point estimates, and the probability of coverage of confidence intervals being lower than the nominal confidence coefficient.
The Drug Evaluation Committee's Data Science Subcommittee's 2022 Continuing Task Force1 focused on adaptive design for case count re-estimation and treatment group selection, investigated statistical inference methods, and prepared a report.
Although the report is intended primarily for statistical analysts involved in planning and conducting clinical trials, examples of clinical trials in which adaptive design has been applied and issues of statistical validity will be of interest to non-statistical members of the Task Force. We hope this report will help you understand adaptive design and conduct appropriate trials.
Published February 2023, revised April 2024
Committee on Drug Evaluation, Japan Pharmaceutical Manufacturers Association
DS Subcommittee 2022 Ongoing Task Force 1
Deliverables of the Drug Evaluation Committee Return to list of all deliverables