The introduction of MedDRA, the Medical Dictionary for Regulatory Activities, as a standardised terminology may have a major impact on the performance of risk management. Thus, MedDRA is likely to have an important effect on the analysis of clinical trial safety data. Review of the most commonly used terms in clinical trial tables from the labelling of ten products indicated that each adverse event could be represented by many MedDRA preferred terms; this might theoretically lead to failure to identify differences in adverse event incidence between treatment arms. Possible solutions are proposed. The use of MedDRA in spontaneous reporting systems is a regulatory requirement in some countries. Variability in modes of implementation and use of the terminology are discussed; these may impose additional limitations on any use of spontaneous data for comparative purposes. There are important differences in the ways that safety databases interface with MedDRA and uncertainty about the most appropriate way to manage version changes. The characteristics of MedDRA must be taken into account when establishing methods for signal detection and its use will affect the retrieval of similar cases as required for signal evaluation. The use of MedDRA in the periodic safety update report is discussed. The possible use of MedDRA in pharmacoepidemiology is highly relevant to risk management, and some issues are briefly outlined. With regard to communication of risk, if MedDRA is introduced into existing product labelling, care must be taken that the change itself does not cause misunderstanding; the most appropriate use of MedDRA in this regard remains to be determined. There is a need for careful evaluation of MedDRA in fulfilling its various functions in pharmacovigilance, followed by definitive regulatory guidance on its use.