Case-control studies in the assessment of drug safety

In a case-control study, the basic approach is to identify a group of subjects newly presenting with the illness of interest and to discover how many of them were exposed to the suspected cause. The frequency of exposure in these cases is then compared with that in a series of control subjects free from the illness under investigation. Many of the epidemiological studies of drug-induced illness now appearing in the literature are of the case-control type, principally because such studies are usually easy to set up, are reasonably cheap, and can often be completed in a short space of time. This paper considers the main areas of methodological importance in carrying out a case-control study of the effects of drugs--i.e. selection of cases, selection of controls, identification of exposure to drugs under study, influence of confounding factors, analysis of the data and interpretation of the results. It is noted that accurate, up to date and accessible registers of disease events and drug exposures relating to identifiable individuals are of great assistance in the conduct of case-control studies.