Test and estimation in binary data analysis under an incomplete block crossover design

Abstract

Under a random effects logistic regression model, we compare two experimental treatments with a placebo in dichotomous data under an incomplete block crossover trial. We develop procedures for testing non-equality of treatments, as well as interval estimators for the relative treatment effects. We employ Monte Carlo simulations to evaluate the performance of these test procedures and interval estimators in a variety of situations. Finally, we use the data taken as a part of the crossover trial that compared the low and high doses of an analgesic with a placebo for the relief of pain in primary dysmenorrhea to illustrate the use of these test procedures and estimators developed here.