Mining for adverse drug events with formal concept analysis.

Alexander Estacio-Moreno (1) , Yannick Toussaint (1) , Cédric Bousquet (2)

1 ORPAILLEUR - Knowledge representation, reasonning
2 SSPIM - Service de santé publique et d'information médicale [CHU Saint-Etienne]

The pharmacovigilance databases consist of several case reports involving drugs and adverse events (AEs). Some methods are applied consistently to highlight all signals, i.e. all statistically significant associations between a drug and an AE. These methods are appropriate for verification of more complex relationships involving one or several drug(s) and AE(s) (e.g; syndromes or interactions) but do not address the identification of them. We propose a method for the extraction of these relationships based on Formal Concept Analysis (FCA) associated with disproportionality measures. This method identifies all sets of drugs and AEs which are potential signals, syndromes or interactions. Compared to a previous experience of disproportionality analysis without FCA, the addition of FCA was more efficient for identifying false positives related to concomitant drugs.

Domaines: Intelligence artificielle [cs.AI]; Sciences pharmaceutiques

Alexander Estacio-Moreno, Yannick Toussaint, Cédric Bousquet. Mining for adverse drug events with formal concept analysis.. Studies in Health Technology and Informatics, 2008, 136, pp.803-808. ⟨hal-00355978⟩