Bayesian analyses of two-treatment crossover studies

Abstract

This review covers Bayesian analyses of two-treatment crossover studies. The main emphasis is placed on the analysis of continuous data although we also consider dichotomous responses. Studies discussed include (1) the simple two-period, two-treatment crossover, (2) the two-period crossover with one baseline measurement, (3) the two-period crossover with two baselines and (4) the extra-period design with two treatment sequences.