Drug Evaluation Committee Flexible Survival Time Modeling Methodology and Extrapolation of Survival Time Using External Information
Data Science Subcommittee
June 2023
The National Institute for Health and Care Excellence (NICE) in the UK, an early adopter of cost-effectiveness evaluation, has published 22 Technical Support Documents (TSDs) on statistical methods for health economic evaluation. These TSDs will be published in 2019. These TSDs are very useful documents for Japan, which will formally introduce cost-effectiveness evaluation in 2019.
The Data Science Subcommittee Continuing Task Force 7 in 2022 is organizing the topics covered in the NICE TSDs and providing information that may be useful for practicing cost-effectiveness evaluation in Japan. This report summarizes flexible survival time modeling and extrapolation of survival time using external information with examples, referring to TSD21 "Flexible methods for survival analysis" published in 2020.
In the past, parametric models using standard distributions such as exponential and Weibull distributions were sufficient for survival time modeling, but recently, with the advent of immunotherapy and other therapies, survival time curves have increasingly shown complex shapes due to delayed response to treatment and the presence of long-term survivors. The survival time curves are now increasingly showing a complex shape. In order to represent such complex shapes, conventional standard models are not sufficient, and more flexible survival time models are needed. Extrapolation beyond the follow-up period of a clinical trial also requires the consideration of external data. In this report, we introduce these methodologies and also present specific examples to help the reader visualize the points to keep in mind when performing modeling.
We hope that this report will help you to select an appropriate method and perform the analysis when you perform survival time modeling or extrapolation using external data.
Japan Pharmaceutical Manufacturers Association Drug Evaluation Committee
Data Science Subcommittee 2022 Ongoing Task Force 7