Hospital admissions resulting from an adverse drug reaction have been studi
ed in the emergency unit of the university hospital in Poitiers during a 27
-day period. This prospective study consisted in documenting all observatio
ns considered as an ADR by the medical practitioner in charge of the patien
t. There were 1235 hospital admissions to the emergency unit during the stu
dy period. Thirty-one (2.5 per cent) of admissions were considered to be dr
ug-related. Women were more often affected than men. Patients with ADR were
classified taking into account the type of pathology and the drug responsi
ble for the effect. Dermatological and gastrointestinal reactions were pred
ominant. Antibiotic and analgesic drugs were the most common drug groups im
plicated in causing an ADR.