It is currently impossible to accurately predict which new drugs will be associated with a significant incidence of idiosyncratic drug reactions and this introduces a significant degree of uncertainty into the drug development process. In the absence of a better understanding of the mechanisms of these reactions, there are a few screening procedures that would likely reduce the probability that a new drug will be associated with idiosyncratic drug reactions. One method is to screen candidates for the formation of reactive metabolites and halt development of drugs that form significant amounts of such metabolites. However, such metabolites are not easy to screen for, and it would also eliminate many candidates that would have been safe if developed. Another risk factor may be the ability of the reactive metabolite to cause cell damage. Even though idiosyncratic reactions appear to be immune-mediated, reactive metabolites that cause cell damage will likely increase the probability of an immune response. Simply developing more potent drugs is likely to decrease the incidence of idiosyncratic drug reactions and it appears that drugs given at a dose of 10 mg/day or less, are associated with a low incidence of idiosyncratic drug reactions. Although the general use of these methods would probably lead to safer drugs, they are far from satisfactory. The use of genomics to search for patterns of change in gene expression that are associated with drugs that cause idiosyncratic reactions has the potential to lead to a more effective screen, but this is unlikely to be a simple process. Ultimately, it is likely that a much better understanding of the mechanisms involved in such reactions will be required to make real progress.