Abstract
In addition to extensive preclinical testing, commercialization of medical devices in the United States under some circumstances may be supported with clinical data derived from single arm studies where the clinical endpoint is compared with a performance goal (PG). Since the performance is expected to be different between two clinically defined mutually exclusive patient subpopulations, the use of weighted PG is often adopted. In this article, we discuss the limitations of the current practice where weights need to be prespecified and fixed, and we propose an alternative approach that relaxes the fixed weight requirements.
Keywords:
Supplementary Materials
Computer codes for generating the simulation results in the article can be found in the supplementary materials
Acknowledgments
We would like to thank Dr. Vandana Bhatia for her English editing, and the anonymous reviewer and associate editor whose comments greatly helped improving this article.
Disclaimer
No official support or endorsement by the Food and Drug Administration of this article is intended or should be inferred.