PATIENT DRUG ATTRIBUTIONS AND POSTMARKETING SURVEILLANCE

Although studies have shown that patients can distinguish probable adv erse drug reactions (ADRs) from adverse clinical events (ACEs) caused by other factors, it is not known whether these attribution judgments add any independent validity to other accepted methods of identifying ADRs, such as physician assessments or epidemiologic data. Data from 2 487 patients receiving fluoxetine and 815 receiving trazodone were use d to see whether such information was redundant when added to standard statistical analysis directed toward detecting ADRs. Relative risk va lues for 14 trazodone or fluoxetine ADRs were selected because each wa s significantly identified by an innovative postmarketing surveillance system. In one analysis, all patient reports were used to compute rel ative risk; in the other, only reports attributed by patients to the s tudy drug were included. Results indicate that taking into account pat ient attribution judgments results in a consistent, albeit modest, inc rease in the discriminatory power of this monitoring method.