The quality of published adverse drug event reports.

Abstract

Case reports of adverse drug events (ADEs) are an important source of information. To determine what variables are reported in ADE case reports, how causality was assessed for each report, and what criteria are specified for publishing ADE case reports. A descriptive analysis of highly significant ADE case reports published in English over a 20-year period was performed. Main outcome measures included frequency distributions for the types of variables reported. The presence of causality assessment and the criteria for submitting an ADE report to practitioner journals were also examined. A highly significant ADE was described in 1520 published case reports during the study period. Three patient variables were reported >90% of the time, while 12 others were reported <25% of the time. Only 1 drug variable was reported >90% of the time; 6 others were reported 14-74% of the time. Most of the relevant ADE variables were reported most often. Added information for drug interactions, medication errors, and allergic drug reactions were reported 61-99% of the time. Less than 1% of ADE reporters objectively assessed the probability of the ADE. All but one journal publishing the most ADE reports did not require such assessment. Professional journals might consider stricter requirements for publishing ADE reports. As a minimum, requirements should include an objective assessment of ADE causality, with explicit recognition in the published text and abstract of the report.