Postulates for the evaluation of adverse reactions to drugs.

Abstract

In most large clinical trials, any unexplained adverse event that occurs during the administration of the study drug is considered to be possibly drug-related. While this approach is effective in identifying the adverse effects of a drug, it does not take into account other, non-drug-related factors that may have caused the event. In very few of these clinical trials has there been an attempt to validate that these events are actually drug-induced; yet, they might still be reported as a side effect in the leaflet insert when the drug is later marketed. When such adverse events are erroneously attributed to drug toxicity, allergy, or idiosyncrasy, it could affect the clinician's decision to use that drug. Some patients may therefore be deprived of a valuable therapeutic tool. Certain postulates should be answered before an adverse effect is considered to be drug-related. It should also be determined whether or not a reaction is due to drug toxicity rather than to allergy or idiosyncrasy. While all adverse events cannot be put to the same stringent tests, under reasonable and safe circumstances, such practices will allow for a more accurate evaluation of a drug and its suspected adverse effects.