Adverse drug reactions (ADRs) have, so far, been classified only on the basis of properties of the drug in question — that is, on the known pharmacology and the dose-dependence of the drug effects — although other factors, such as the delay in occurrence of the ADR and failure of therapy, are also sometimes taken into account.

But Jeffrey Aronson and Robin Ferner argue in the British Medical Journal that this approach gives limited insight into ADRs and that a comprehensive classification should take into account additional factors that relate to the drug, the patient and the reaction itself. For example, corticosteroid-related osteoporosis depends on the dose and the duration of treatment, and some reactions, such as asthma caused by β-adrenoceptor antagonists, do not occur in all patients, who have differing susceptibilities.

Aronson and Ferner therefore propose a three-dimensional classification system called DoTS, based on dose relatedness (Do), the time course of the reaction (T) and the susceptibility (S) of the patient. Dose relatedness is defined in terms of reactions that occur at supra-therapeutic doses (toxic reactions), standard therapeutic doses (collateral reactions) and sub-therapeutic doses (hyper-susceptibility reactions). Timing is divided into time-dependent and time-independent effects: time-independent reactions occur at any time during a treatment course, whereas time-dependent reactions can occur at six stages — rapid, first dose, early, intermediate, late (including withdrawal reactions) and delayed. Susceptibility factors alter the risks of ADRs in individuals, and include age, sex, physiological and genetic variation, and drug interactions.

Three examples of DoTS classifications of ADRs are given. For instance, the classification of osteoporosis due to corticosteroids is: dose-relatedness, collateral; time-course, late; suceptibility factors, age and sex. A more sophisticated analysis is also possible, which requires an estimate of the probability of an ADR at different doses and times after administration for different degrees of susceptibility, and can be displayed as a series of three-dimensional graphs or nomograms. The authors conclude that their proposed classification “should provide important insights for drug development and regulation, for pharmacovigilance, for monitoring patients, and for the prevention, diagnosis, and treatment of adverse drug reactions.”