Postmarketing Follow-up

THERE is a general but strongly held view that, after drugs are marketed, additional epidemiologic information is needed on their clinical effects, particularly adverse effects. This notion is based soundly on the limited size and scope of the clinical experience with a drug before marketing. Recently, considerable interest has been evident in this area and has focused on the term "postmarketing surveillance" (PMS). That the interest is widespread and intense is reflected in the many meetings that have taken place, articles that have been written, and committees that have been formed to consider the designing of large formal systems of PMS.^1-15 Given the diffuse nature of the subject, it is essential to take an objective and informed look at the legitimate needs in this area and to consider practical means of satisfying these needs. Without a clear understanding of the nature of the problems that can be addressed epidemiologically