Computerized surveillance of adverse drug reactions in hospital: Implementation

Objective: To implement and measure the effects of automatic computerized laboratory signals (ALS) as a detection support tool of adverse drug reactions (ADRs) in hospital. Methods: This was a prospective observational study of a total of 192 patients (199 sequential medical admissions) during a 2-month period in a 34-bed medical ward at the Hadassah University Hospital, Jerusalem, Israel. The study involved the routine (daily) distribution to staff physicians of lists of automatic signals generated from computerized laboratory data as potential indicators of ADRs. Patient charts were reviewed by the clinical pharmacology team for ADRs and to see whether these were recognized by the staff physicians. Results: Seventy-one ADRs were detected in 64 of the 199 (32%) admissions. Twenty-seven per cent of the ADRs were serious, 9% of the admissions were due to ADRs. Two hundred and ninety-five ALS were generated involving 69% of the admissions. Sixty-one per cent of the ADRs were identified by ALS. ALS were present in 58% of the ADR negative admissions. Eighty-five per cent of the ADRs were recognized as such and 19% of the ALS-positive ADRs were not recognized by the staff physicians. Conclusions: The routine implementation of ALS doubled the number of ADRs recognized by the physicians while patients were hospitalized in the medical ward. The use of the system appeared valid, simple and potentially cost-effective. Received: 17 June 1998 / Accepted in revised form: 1 October 1998