Drug Evaluation Committee Statistical Methods for Estimating Treatment Effects Using Non-randomized Controlled Trial Data
Data Science Subcommittee
February 2020
From April 2019, cost-effectiveness evaluation was institutionalized as a system to complement the existing drug price standard system. Under this system, a company analysis is conducted after a pre-analysis consultation on analysis methods, etc., and after the scientific validity of the company analysis is verified (reviewed) by a public analysis team, a comprehensive evaluation is conducted by a specialized organization for cost-effectiveness evaluation, and the usefulness-based additional portion (and operating income in case of cost accounting system) of the drug price is adjusted based on the incremental cost-effectiveness ratio. This is a system in which the incremental cost-effectiveness ratio is used as the basis for adjusting the additional utility portion of the drug price (and operating income in the case of the cost accounting method). In the process of conducting cost-effectiveness analysis under this institutionalization, it is necessary to estimate the treatment effect of a new drug treatment on a comparator treatment. If randomized controlled trial data are not available for estimating treatment effects, it may be useful to use data from non-randomized trials, including observational studies (non-randomized controlled trials). The DS Subcommittee will continue to work on the use of observational data to inform the DS Subcommittee in T5 of Technical Support Document 17: The use of observational data to inform treatment effect estimation, published by NICE (UK). The use of observational data to inform estimates of treatment effectiveness in technology appraisal: methods for comparative individual patient data (2015) published by NICE in the U.K., we prepared a report describing the analytical methods used in the analysis of data from nonrandomized controlled trials, including propensity score-based methods, control variable methods, and regression discontinuity designs. This report focuses on analytical methods that can be applied using standard statistical software, and we hope that it will be a useful reference for those who are actually practicing cost-effectiveness analysis in the future. The references to/references to TSD 17 in this report are the interpretation of this task force, and NICE is not responsible for any of the references. In addition, not all of TSD 17 has been introduced, but some excerpts have been extracted and introduced. In introducing them, the Task Force has added its own interpretations and additional explanations. These are the Task Force's own summaries and do not represent the views of NICE.
Deliverables of the Drug Evaluation Committee Return to list of all deliverables