Pharmacoepi
Analysis of complex case-control studies in pharmacoepidemiology
Funding: funded as part of FOR 534 up to April 2007
Description
Severe cutaneous adverse reactions like Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are rare but life-threatening disorders mainly caused by drugs. At the moment a multitude of drugs are suspected to cause them. For assessing their risk, the two international case-control studies SCAR and EuroSCAR were performed between 1989-1995 and 1997-2001, respectively. Since the pathophysiology as well as further risk factors of SJS and TEN are mainly unknown, the third international case-control study RegiSCAR (funded by the European Community) focuses on genetic and immunologic questions. For this reason biological samples from cases and controls were centrally stored. Although the question concerning drug risk is still interesting, the controls included in RegiSCAR cannot be used to answer this. Recruitment of cases and controls for the RegiSCAR study was completed last year. In this project, the methodological problems of these complex studies are considered. This includes careful model building in the case-control studies SCAR and EuroSCAR and the case series RegiSCAR to identify a list of candidate drugs. For case series the case-crossover method is assessed and will be adapted to the underlying situation where time-dependent effects appear. Furthermore, the use of external information for the analysis of case series is investigated. Finally, high-dimensional genomic data will be analyzed to ¯nd genetical risk factors. Recently proposed methods will be adopted in order to take the correlation structure of the data and the magnitude of the effects into account.
Primary investigator
Prof. Dr. Martin Schumacher (IMBI)