Background: For the purposes of drug approval, the interchangeability of a generic drug and the corresponding brand-name drug is based on the criterion of "essential similarity," which requires that the generic drug have the same amount and type of active principle, the same route of administration, and the same therapeutic effectiveness as the original drug, as demonstrated by a bioequivalence study However, bioequivalence and therapeutic effectiveness are not necessarily the same. Objective: This review summarizes available data comparing the bioequivalence and therapeutic efficacy of brand-name psychoactive drugs with those of the corresponding generic products. Methods: Relevant information was identified through searches of MEDLINE, Current Contents/Clinical Medicine, and EMBASE for English-language articles and English abstracts of articles in other languages published between 1975 and the present. The search terms used were generic drug, branded drug, safety, toxicity, adverse events, clinical efficacy, bioequivalence, bioavailability, psychoactive drugs, and excipients. Results: Few publications compared the bioequivalence and efficacy of brand-name and generic psychoactive drugs. Those that were identified revealed differences in the efficacy and tolerability of brand-name and generic psychoactive drugs that had not been noted in the original bioequivalence studies. Specifically, study found that plasma levels of phenytoin were 31% lower after a switch from a brand-name to a generic product. Several controlled studies of carbamazepine showed a recurrence of convulsions after the shift to a generic formulation. After a sudden recurrence of seizures when generic valproic acid was substituted for the brand-name product, an investigation by the US Food and Drug Administration found a difference in bioavailability between the 2 formulations. Statistically significant differences in pharmacokinetic variables have been reported in favor of brand-name versus generic diazepam (P < 0.001). Finally, a case report involving a double-shift between brand-name paroxetine hydrochloride and generic paroxetine mesylate cast doubt on the tolerability and efficacy of the generic formulation. Conclusion: The essential-similarity requirement should be extended to include more rigorous analyses of tolerability and efficacy in actual patients as well as in healthy subjects. Copyright (C) 2003 Excerpta Medica, Inc.
