We consider the outcomes of a clinical trial as determined by one, or sever
al, possibly hidden causes. This paper proposes a statistical model that al
lows such a distinction of causes not only for the main, or therapeutic, ef
fects 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 effe
ct has occurred or not. A case study provides an example of the way this mo
del can help to solve some problems of suspected drug toxicity. Finally, th
e model is shown to be a part of a hierarchy of models and the way to selec
t a best model is investigated. Copyright (C) 2001 John Wiley & Sons, Ltd.