Pharmacoepidemiology and its input to pharmacovigilance.

Summary The aim of this article is to discuss the contribution of pharmacoepidemiology to pharmacovigilance in the context of drug safety monitoring. Because this contribution requires a comparison of the approaches and a presentation of their respective merits, we have tried to show that this comparison did not necessarily mean opposing these methods, which would not serve any useful purpose. We wanted to underline the fact that the notion of contribution of one to the other implied a complementarity that we propose to demonstrate. Pharmacovigilance is essentially based on the study of spontaneous reports and the clinical and pharmaceutical analysis of the implication of given drugs in adverse drug reactions. It yields performances that are unequalled to this day when safety warning signs relate to rare adverse events or to drugs that are rarely used, or when the drug-related event has specificities that are not found for other aetiologies. Pharmacoepidemiology for its part, through its population-related studies and the possible use of a comparison group, enables a quantification of risks that is impossible to carry out using spontaneous reporting alone. It also enables the highlighting and quantification of safety signals for events that are frequent and multifactorial, where the role of a drug in individual cases can be difficult to detect, and therefore difficult to identify from spontaneous reporting. All in all, this is the major contribution of pharmacoepidemiology to pharmacovigilance, where it can be said that it complements pharmacovigilance in drug safety monitoring but in no way could replace it.