We consider the outcomes of a clinical trial as determined by one, or several, possibly hidden causes. This paper proposes a statistical model that allows such a distinction of causes not only for the main, or therapeutic, effects but also for the side, or toxic, effects. More specifically, we focus on trials where the effects are naturally dichotomized, that is, where the health of a patient has improved or not, and where a specific adverse effect has occurred or not. A case study provides an example of the way this model can help to solve some problems of suspected drug toxicity. Finally, the model is shown to be a part of a hierarchy of models and the way to select a best model is investigated.
Copyright 2001 John Wiley & Sons, Ltd.