Postmarketing surveillance: accuracy of patient drug attribution judgments.

Data from two samples of ambulatory patients participating in a postmarketing surveillance study, one receiving antibiotics and another receiving tricyclic antidepressant agents, are presented, indicating that patients appear to be capable of correctly discriminating probable adverse drug reactions from other adverse clinical events. However, attribution accuracy depended both on the surveillance method and on how reports of the adverse clinical events were obtained. Discrimination was better when patients were reporting adverse clinical events spontaneously than when the interviewer probed for recall in a systematic inquiry. Discrimination was also better when the adverse clinical events were obtained from a staff-initiated surveillance method than from a self-monitoring, patient-initiated telephone-reporting method--probably because the latter method generates an excessively strong tendency to report mainly those adverse clinical events suspected of being drug related.