Answers are grounded in this paper and cite their sources.
Methodological analysis demonstrates optimal sample sizes across treatment selection strategies for stratified patient cohorts, highlighting conditions where Bayesian methods outperform pooling.
Various strategies are investigated for selecting one of two medical treatments when patients may be divided into k ≥ 2 categories on the basis of their expected differences in response to the two treatments. The strategies compared are (1) k independent decisions, (2) a single overall decision made on the basis of simple pooling, and (3) the Bayes strategy. The optimal clinical trial size is supplied for each strategy, and conditions delineated under which each of the strategies is to be preferred.
No takes yet. Share an insight, caveat, or question.
Allan Donner (1979) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: