Assessing causality in clinical trials, A Sure Outcome of Random Events (SORE) Model

This paper considers causal assessment in the case of a particular model, namely a Sure Outcome of Random Events (SORE) model developed for the analysis of data from a randomized controlled trial of a drug. A distinctive feature of this model is that it takes into account different observable effects and for each of these, different possible latent causes. In particular, the model distinguishes between two kinds of observable effects, a therapeutic effect and a side-effect. For each observable effect, two latent factors are considered, i.e. a pharmacological (or explained) factor and a residual (or unexplained) one. The SORE model allows one to distinguish naïve causal assessment, relying on a prima facie analysis of the empirical distribution of the observable variables, from a causal attribution based on a structural model taking into account both observable and latent variables. This approach is illustrated by a numerical example, along with a case study based on real data.