|
Abstract:
|
In rare disease drug development, difficulty in enrollment can make the conduct of randomized controlled trials (RCTs) infeasible because of their size, duration, cost, patient preference, or in some cases ethical constraints. One way to mitigate recruitment challenges and inadequate sample size issue is to introduce data-driven priors when borrowing data from external sources, that is, to apply Bayesian methodology. A crucial consideration when we apply data-driven priors vs. objective priors is that we need some assurance of the consistency, objectivity and comprehensiveness of the synthesized data. To achieve this, propensity score matching methods can be used to control for confounding by matching experimental subjects and control subjects on a set of pre-treatment characteristics. In this talk, applications based on real clinical trial data will be discussed.
|
