May 7, 1997

Adverse Drug Events in Hospitalized Patients-Reply

Author Affiliations

Brigham and Women's Hospital Boston, Mass

JAMA. 1997;277(17):1353-1354. doi:10.1001/jama.1997.03540410031018

In Reply.  —Dr Davis expresses his frustration with the ADR reporting system at his hospital. He has difficulty determining whether individual events are legitimate ADRs and resents being required to report them, particularly for the marginal reactions. We have several comments. We prefer the term "adverse drug event" to "adverse drug reaction," because ADRs have generally excluded injuries to patients that occurred as the result of a medication error, which are the most clinically important of these injuries. Next, it is clear that spontaneous reporting identifies only a very small fraction of ADEs.1 We think that hospitals should use computerized approaches to identify these events2 and that ongoing monitoring is essential; such computer monitoring can increase the number of events found by as much as 20-fold.1 Also, whether a specific event does represent an ADE can be reliably determined by experienced reviewers (k=0.98 in our study); we only consider