|
Abstract:
|
Existing statutes require manufacturers to demonstrate evidence of effectiveness through the conduct of adequate and well-controlled studies to obtain marketing approval of a therapeutic product. What constitutes adequate and well-controlled studies is usually interpreted as randomized controlled trials (RCTs). However, these trials are sometimes unfeasible because of their size, duration, cost, or ethical concerns. One way to reduce sample size requirements is to borrow information through the use of a prior, i.e., to apply the Bayesian methodology. In an era of data transparency, data derived from external control can complement information provided by RCTs by using it as a prior. One important consideration when forming data-driven prior is that there must be some form of consistency of the data, i.e., exchangeability. For this reason, propensity score matching methods can be used to control for confounding by matching the new treatment and control units based on a set of measured covariates. In particular, a matching scheme based on propensity scores estimated through generalized boosting, are explored and applied to a disproportionately randomized clinical trial example.
|
