|
Abstract:
|
Clinical investigations best serve the interests of public health through the use of efficient trial designs comprised of interim monitoring to preserve ethical, scientific, and economic concerns. In this talk, we will describe recent statistical methodology in clinical trial designs that are adaptive, allowing flexibility for sample size determination while maintaining scientific validity. The methods include the design and inference of continuous outcomes, censored survival data with staggered entry, and correlated outcomes with covariates. Important theoretical properties are investigated. Moreover, adaptive designs within the framework of a Bayesian decision theoretical approach will be discussed. The proposed designs and corresponding estimation methods are applicable to various areas of health research, especially cancer clinical trials.
|
