Multieentre Clinical Trials and Variance Components

Abstract
A method of analysing regression data is presented for a model in which the dependent variable has a binomial distribution and the regression contains both fixed and random components. The method is then used to analyse treatment success data for a multicentre clinical trial study.