Pharmacotherapy is often the single most important medical intervention in the care of the elderly. However, there are obvious concerns about the vulnerability of this group to adverse drug reactions (ADRs). A rapidly accumulating literature regarding changes in pharmacokinetics and pharmacodynamics with advancing age suggests a strong pharmacologic basis for such concerns. Yet, the results of epidemiologic studies exploring the relationship between age and ADRs are ambiguous. Interpretation of the results of these studies is limited by inconsistent definitions of outcomes of interest and failure to control for important age-related covariates including the clinical status of the patient and the number of medications that a patient is receiving. Some recent studies have investigated age-related aspects of specific adverse consequences of drug therapy. For example, age, in and of itself, does not appear to be a risk factor for bleeding complications of warfarin therapy. Older patients may actually be at less risk than younger patients to experience depression associated with beta-blocker therapy. Although examination of data from premarketing studies might be considered a promising strategy to explore the relationship between age and ADR risk, the small number of truly elderly subjects included in these studies greatly limits their usefulness. Postmarketing studies utilizing databases containing clinical data for large numbers of older patients may provide the optimal approach for investigating whether old age is an independent risk factor for ADRs.