Comparison of adverse drug reactions detected by pharmacy and medical records departments.

Adverse drug reactions (ADRs) detected by the pharmacy and medical records departments of a multispecialty teaching hospital were studied. The charts of all adult patients who were identified by the pharmacy or medical records departments as having had an ADR and who were discharged from the hospital between July and September 1990 were reviewed. Data on patient demographics and the characteristics of the ADRs were collected, and the causality and severity of each ADR were assessed by two pharmacists and one physician. A total of 110 charts representing 117 ADRs were reviewed. Twenty-five (21%) of the ADRs were identified by the pharmacy department and 101 (86%) by the medical records department; 9 (8%) were reported by both departments. The pharmacy and medical records groups of patients were demographically similar, except that the percentage of patients admitted through the emergency room was significantly smaller for the pharmacy department group. ADRs identified by the pharmacy were most commonly cutaneous, and those identified by medical records were most commonly neurologic. For the pharmacy department, hypersensitivity reactions accounted for the largest number of ADRs, while for medical records the largest number involved abnormal laboratory test values. Anti-infectives were involved in two thirds of the pharmacy-identified ADRs, compared with only a fifth of the ADRs identified by medical records. Mean causality and severity scores did not differ significantly between the groups. The medical records department identified four times as many ADRs as the pharmacy department. Observed differences in the number and types of reactions, manifestations, patient locations, and suspected drugs probably reflect the different surveillance methods and ADR definitions used by the two departments.