Emergency department visits caused by adverse drug events: results of a French survey.

BACKGROUND: Adverse drug events (ADEs) are a substantial cause of hospital admissions. However, little is known about the incidence, preventability and severity of ADEs resulting in emergency department visits. To address this issue, we conducted a prospective survey in emergency departments of French public hospitals. METHODS: This study was performed over two periods of 1 week each, one in June 1999 and one in December 1999, in emergency departments of five university hospitals and five general hospitals throughout France. All patients aged>or=15 years presenting with medical complaints were included in the study. Trauma patients, those with gynaecological conditions and those with alcohol intoxication or intentional drug poisoning were excluded from the study. Each patient was assessed by two local emergency physicians to determine whether the visit was the result of an ADE. All medical records were subsequently validated by an independent group of medical lecturers in iatrogenic disorders. RESULTS: Out of a total of 1937 patients consulting, 1562 were taking at least one drug during the previous week and were included for analysis; 328 (21%; 95% CI 19, 23) of these patients consulted an emergency physician because of an ADE. Patients with ADEs were older than those without (mean age 63.5 vs 54.8 years; p<0.0001). Furthermore, ADE patients were more likely to have a higher severity presentation than the non-ADE group (p=0.019). The number of drug exposures was significantly higher in patients with an ADE than in those without (mean number of medications 5.17 vs 3.82; p<0.0001). On multivariate analysis, only age and the number of medications taken were significantly associated with adverse events. In total, 410 drugs were incriminated in the occurrence of 328 ADEs. The most frequently incriminated drug classes were: (i) psychotropic agents (n=84; 20.5%); (ii) diuretics (n=48; 11.7%), anticoagulants (n=38; 9.3%) and other cardiovascular drugs (n=63; 15.4%); and (iii) analgesics, including NSAIDs (n=57; 13.9%). Preventability could be assessed in 280 of the 328 cases. In 106 cases (37.9%), the ADE was judged to be preventable. CONCLUSION: ADEs leading to emergency department visits are frequent, and many are preventable, confirming that there is a need to develop prevention strategies.