[Case-non case studies: Principles, methods, bias and interpretation].

Case-non case studies belongs to the methods assessing drug safety by analyzing the disproportionality of notifications of adverse drug reactions in pharmacovigilance databases. Used for the first time in the 1980s, the last few decades have seen a significant increase in the use of this design. The principle of the case-non case study is to compare drug exposure in cases of a studied adverse reaction with that of cases of other reported adverse reactions and called "non cases". Results are presented in the form of a reporting odds ratio (ROR), the interpretation of which makes it possible to identify drug safety signals. This article describes the principle of the case-non case study, the method of calculating the ROR and its confidence interval, the different modalities of analysis and how to interpret its results with regard to the advantages and limitations of this design.