Investigation of Trigger Tools for Detecting Adverse Drug Reactions

Abstract

Background Despite the high morbidity and mortality associated with adverse drug events (ADEs) and adverse drug reactions (ADRs) the ability to detect and report them systematically remains inefficient. Aim To investigate if the use of a modified ADE trigger tool could improve the identification of ADRs in hospitals. Method Retrospective review of patients' discharge summaries was conducted at a tertiary care facility over a 5‐week period. A modified ADE trigger tool (based on a previous model) was used to assess four high‐risk medicines (warfarin, heparin, morphine, benzodiazepines). ADEs identified were classified using both the Naranjo probability scale and the WHO ADR classification. Results Data from 528 patients were reviewed, of which 286 (54%) received at least 1 of the 4 high‐risk medicines. 45 patients (8.5%) were identified using the modified trigger tool as having had an ADE during their hospital stay. The modified ADE trigger tool improved the identification of ADRs compared with the spontaneous reporting system, regardless of the classification method used. Both the WHO and Naranjo classification scales yielded higher rates than the spontaneous reporting rate of 0.07% recorded by the New Zealand Centre for ADR Monitoring. Conclusion The modified ADE trigger tool appears to be a sensitive method for identifying ADRs. The results indicate that using the modified trigger tool yields significantly higher ADR reporting rates than the traditional spontaneous reporting system.

Adverse drug events among adult inpatients: a meta-analysis of observational studies

“…Most (26) of the articles, except two, 27,28 described how the participants were selected. Six of the articles in the study did not state participant inclusion/exclusion criteria explicitly.…”

“…Six of the articles in the study did not state participant inclusion/exclusion criteria explicitly. [27][28][29][30][31][32] Of the 28 studies included, 13 identified and assessed avoidable events separately. 25,26,[31][32][33][34][35][36][37][38][39][40][41] Three studies restricted the analysis to ADRs, clearly excluding the avoidable events.…”

“…30,[44][45][46] In the remaining studies (8), it is not clear whether such events were included. 27,29,[47][48][49][50][51][52] Nine articles did not define the profile of the team involved in identifying and evaluating the events, 26,27,29,30,[36][37][38]49,50 and only three mentioned team training. 28,44,46 Nearly all the articles, except two, 36,48 described standardization of data collection and ADE case identification procedures.…”

“…In those studies, the rates presented were 4Á4 ADEs per 100 patient-days, 39 3Á6 ADEs per 100 inpatients 44 and 8Á5 ADEs per 100 inpatients. 27 The 25 studies that presented proportion of events were grouped by event identification (SR,RM,PM) method. The summary proportion differs between the subgroups, and the intervals do not overlap.…”

“…27 The 25 studies that presented proportion of events were grouped by event identification (SR,RM,PM) method. The summary proportion differs between the subgroups, and the intervals do not overlap.…”