Drug Evaluation Committee Introduction of Sequential Multiple Assignment Randomized Trials (SMART)
November 2022
Sequential Multiple Assignment Randomized Trials (SMART) are a design for establishing a series of dynamic treatment plans through sequential, multi-stage random assignment. While SMART is currently attracting attention overseas, no Japanese-language texts on the subject currently exist.
Therefore, we—the SMART Subteam of Task Force 4 (Causal Inference) of the Data Science Working Group, Drug Evaluation Committee, Japan Pharmaceutical Manufacturers Association (FY2021)—have created this document to introduce SMART to clinical practitioners using real-world examples and to provide an Explanation of the design overview and estimation methods to statistical experts.We have also included sample data and SAS code so that readers can perform actual analyses and gain a practical understanding of the method.
We hope this document will aid in understanding SMART and assist with the planning, analysis, and interpretation of results in trials focused on dynamic treatment planning.
Japan Pharmaceutical Manufacturers Association Drug Evaluation Committee
, DS Subcommittee, Task Force 4, FY2021