2019 HSR&D/QUERI National Conference
4005 — Adverse Drug Event Reporting among Patients Who Discontinued Empagliflozin or Apixaban in the Veterans Health Administration
Lead/Presenter: Fran Cunningham, VA Center for Medication Safety
All Authors: Aspinall SL (VA Center for Medication Safety, VA Center for Health Equity Research and Promotion), Fina PM (Chicago State University), Zhao X (VA Center for Health Equity Research and Promotion) Glassman PA (VA Center for Medication Safety) Moore VR (VA Center for Medication Safety) Au A (VA Center for Medication Safety) Cunningham FE (VA Center for Medication Safety)
Passive and active surveillance of adverse drug events (ADEs) are used to help inform safe and appropriate medication use. The objective of this project was to examine reporting rates at increasingly "higher" levels of spontaneous reporting systems (i.e., point of care, then national VA healthcare system, and then the FDA) of apixaban and empagliflozin ADEs.
This was a retrospective cohort study of outpatients who discontinued apixaban or empagliflozin within three years of FDA approval. We also used an active surveillance strategy and identified a subset of patients with ICD-9/10 codes associated with a possible ADE (e.g., GI bleed-apixaban, ketoacidosis-empagliflozin). Stratified random samples of charts were reviewed to determine if patients discontinued the medication due to an ADE. Then, we ascertained if ADEs from the charts were entered into the VA electronic health record/point of care reporting system (Adverse Reaction Tracking System [ARTS]), VA national web-based system (VA Adverse Drug Event Reporting System [VA ADERS]), and FDA MedWatch, three spontaneous reporting databases.
From the cohort of 2,973 patients who discontinued apixaban, 321 (10.8%) were randomly sampled for chart review (61 patients with ICD codes for GI bleed and/or intracranial hemorrhage). During chart review, 88 ADEs were identified; 40/61 (65.6%) from the subset with ICD codes. Of the 88 total ADEs, 23.3%, 10.2%, and 6.8% were reported in ARTS, VA ADERS, and FDA MedWatch, respectively. Of the 1,555 patients who discontinued empagliflozin, 179 (11.5%) were randomly sampled for chart review (40 patients with ICD codes for ketoacidosis and/or amputation). During chart review, 78 ADEs were identified; 19/40 (47.5%) from the subset with ICD codes. Of the 78 total ADEs, 28.2%, 19.2%, and 7.7% were reported in ARTS, VA ADERS, and FDA MedWatch, respectively.
Compared with chart documentation, we found substantial underreporting of apixaban and empagliflozin ADEs that became worse at each "higher" level of spontaneous reporting.
Processes could be developed to allow reasons for medication discontinuation to be documented in a point of care reporting system and to capture ICD codes for ADE verification and reporting so the information is available for patient care.