ESTIMATING COSTS OF CLUSTERED OUTPATIENT ADVERSE DRUG REACTIONS WITHIN THE VETERANS HEALTH ADMINISTRATION USING THE VA ADVERSE DRUG EVENT REPORTING SYSTEM
Author(s)
Vu M1, Moore V2, Au A2, Bounthavong M3, Aspinall SL4, Good CB5, Glassman P6, Cunningham F2
1Center for Medication Safety (VA MedSAFE) and VA Center for Health Equity Research and Promotion, Stone Mountain, GA, USA, 2Veterans Health Administration Center for Medication Safety (VA MedSAFE), Pharmacy Benefit Services, Hines, IL, USA, 3Veterans Health Administration Academic Detailing Program Office, San Diego, CA, USA, 4Center for Medication Safety (VA MedSAFE) and VA Center for Health Equity Research and Promotion, Pittsburgh, PA, USA, 5Veterans Health Administration Center for Medication Safety (VA MedSAFE), University of Pittsburgh Medical Center, Pittsburgh, PA, USA, 6Veterans Health Administration Center for Medication Safety (VA MedSAFE), VA Greater Los Angeles Healthcare System, Hines, IL, USA
OBJECTIVES: Surveillance of adverse drug reactions (ADRs) within healthcare systems is important to patient safety. ADRs may include multiple drug-related symptoms (symptom clusters), and severe ADRs require hospitalization/urgent intervention. There have been no cost estimates of ADRs within the Veterans Health Administration (VA) using its Adverse Drug Event Reporting System (ADERS). This quality improvement project aims to estimate direct medical costs associated with severe ADRs and compare costs of single-symptom and symptom-clustered cases. METHODS: Outpatient ADRs reported in VA-ADERS as severe, from fiscal years 2014 to 2018, were analyzed. ADR symptoms were assessed as Medical Dictionary for Regulatory Activities (MedDRA) Preferred Terms and grouped by the primary suspected drug class. Hospitalization outcomes came from VA-ADERS reports. Cases without event dates or reported >365 days after occurrence or treated at non-VA facilities were excluded. MedDRA Preferred Terms were mapped to ICD9/10 codes and assigned hospitalization aggregated costs from the Healthcare Cost and Utilization Project’s Nationwide Inpatient Sample. For ADRs with symptom clusters, a hierarchical system, where one symptom was assigned as the primary symptom and driver of healthcare utilization, was developed based on expert review. RESULTS: Of 8,991 ADR cases, 3,893 (43.3%) had clusters and 6,712 (75.3%) involved hospitalization. The drug classes with the highest ADR costs were anticoagulants ($17.2M), ace inhibitors ($5.9M), and non-steroidal anti-inflammatory drugs ($2.1M). The most common primary symptoms clustered included gastrointestinal hemorrhage, angioedema, and acute kidney injury. Drug classes most commonly paired with symptom clusters included anticoagulants, ace inhibitors, and antineoplastics. Average cost of ADRs with symptom clusters were lower compared to those in non-clustered cases ($8,900 vs. $9,275). CONCLUSIONS: ADR reporting systems can be tools for assessing economic burden. A large proportion of ADRs included symptom clusters, requiring methods to improve cost-estimation. Results will be used to help inform medication safety policies and initiatives.
Conference/Value in Health Info
2019-05, ISPOR 2019, New Orleans, LA, USA
Value in Health, Volume 22, Issue S1 (2019 May)
Code
PDG31
Topic
Economic Evaluation, Epidemiology & Public Health
Topic Subcategory
Cost/Cost of Illness/Resource Use Studies, Safety & Pharmacoepidemiology
Disease
Drugs, No Specific Disease
Explore Related HEOR by Topic
