Background: 'Preventability' is a crucial concept in the literature on adverse drug effects. However, a systematic review of the definitions of preventability of adverse drug effects has suggested that none fits all circumstances. Furthermore, when the reliability of these definitions has been examined they have been found to be imperfect. Objective: To propose and outline a method for determining the theoretical preventability of an adverse drug effect, based on frameworks for classifying adverse drug reactions - the EIDOS and DoTS methods. Methods: EIDOS is based on the mechanism of action of the drug. It observes that a drug (an Extrinsic species) causes an adverse effect by interacting with an Intrinsic species that is its target when the two are Distributed together, and that the resulting pathophysiological Outcome (the adverse effect) causes the Sequela (the adverse reaction). DoTS observes that the Dose-relatedness of the adverse effect compared with the beneficial effect is relevant (determining toxic, collateral, or hypersusceptibility effects), that adverse effects have Time-courses (varying from immediate to delayed), and that there are individual Susceptibility factors. Results and Discussion: We have elicited many published examples that show that each of these factors in the causation of an adverse drug effect can be adduced to assess its preventability. We have constructed a flowchart that illustrates how the processes can be logically analysed. Conclusions: This approach suggests methods for devising prospective preventive strategies and for deciding retrospectively whether an adverse reaction in an individual should have been prevented.