Comparative study of prospective surveillance and voluntary reporting in determining the incidence of adverse drug reactions.

The results of a hospitalwide voluntary reporting program for adverse drug reactions (ADRs) was compared with the results of a short-term, intensive, prospective surveillance program conducted by a pharmacist on a medical and a surgical patient care unit of a large teaching hospital. Data generated by the voluntary system were collected for a 45-day period; for the prospective surveillance, data were collected for 21 days. Strict definitions on the categories of probability, severity and mechanism of adverse reactions were employed. The incidence of definite and probable ADRs in the voluntary system was 0.08%; for prospective surveillance, the incidence was 7.2% (5.9% of 85 surgical patients and 9.0% of 67 medical patients). The incidence of ADRs in the prospective study, based on patient-drug exposures, was 1.0%. In the prospective study, patients experiencing ADRs received significantly more medications than those not experiencing ADRs. The operation of an ADR monitoring and reporting system for the purpose of maintaining incidence data was not judged to be cost effective. Periodic prospective surveillance programs on representative patient population samples may be valuable in determining true incidence figures.