Drug Evaluation Committee Causal Reasoning for Understanding ICH E9(R1) - Time Dependent Treatment
September 2022
The various causal inference methods discussed in “Causal Inference to Aid in Understanding ICH E9(R1),” published in July 2022, were based on the assumption of time-fixed treatment.However, in practice, it is often assumed that there is interest in causal comparisons that include time-varying treatments. Furthermore, we believe that understanding the concepts of causal inference for time-varying treatments is helpful when defining treatment regimens that include such treatments and, by extension, when setting the associated estimands.
Therefore, we—the Subteam on Estimation of Time-Varying Treatments Using Causal Inference, part of the FY2022 Continuing Task Force 4 of the Data Science Working Group under the Drug Evaluation Committee of the Japan Pharmaceutical Manufacturers Association—have explained in this deliverable the basic concepts of causal inference for time-varying treatments, various estimation methods, application examples, and SAS implementation examples.
We hope this deliverable will aid in understanding ICH E9(R1) and assist in the planning, analysis, and interpretation of results in clinical trials and observational studies focused on treatment regimens, including time-dependent treatments.
Japan Pharmaceutical Manufacturers Association (JPMA) Drug Evaluation Committee,
DS Subcommittee, FY2022 Continuing Task Force 4