Premium
Bayesian approach to the design and analysis of non‐inferiority trials for anti‐infective products
Author(s) -
Gamalo Meg A.,
Tiwari Ram C.,
LaVange Lisa M.
Publication year - 2013
Publication title -
pharmaceutical statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.421
H-Index - 38
eISSN - 1539-1612
pISSN - 1539-1604
DOI - 10.1002/pst.1588
Subject(s) - frequentist inference , bayesian probability , sample size determination , computer science , observational study , econometrics , prior probability , margin (machine learning) , statistics , machine learning , bayesian inference , artificial intelligence , mathematics
In the absence of placebo‐controlled trials, determining the non‐inferiority (NI) margin for comparing an experimental treatment with an active comparator is based on carefully selected well‐controlled historical clinical trials. With this approach, information on the effect of the active comparator from other sources including observational studies and early phase trials is usually ignored because of the need to maintain active comparator effect across trials. This may lead to conservative estimates of the margin that translate into larger sample‐size requirements for the design and subsequent frequentist analysis, longer trial durations, and higher drug development costs. In this article, we provide methodological approaches to determine NI margins that can utilize all relevant historical data through a novel power adjusted Bayesian meta‐analysis, with Dirichlet process priors, that puts ordered weights on the amount of information a set of data contributes. We also provide a Bayesian decision rule for the non‐inferiority analysis that is based on a broader use of available prior information and a sample‐size determination that is based on this Bayesian decision rule. Finally, the methodology is illustrated through several examples. Published 2013. This article is a U.S. Government work and is in the public domain in the USA.