Related Experiment Videos
Using computerized data to identify adverse drug events in outpatients.
B Honigman1, J Lee, J Rothschild
1University of Colorado Health Sciences Center, Denver, USA. benjamin.honigman@uchsc.edu
Summary
Computerized programs effectively detect adverse drug events (ADEs) in ambulatory care, with free-text searches proving particularly valuable. These ADEs are common, sometimes requiring hospitalization, highlighting the need for such detection systems.
Area of Science:
- Pharmacovigilance
- Health Informatics
- Clinical Decision Support
Background:
- Adverse drug events (ADEs) represent a significant patient safety concern in ambulatory settings.
- Current methods for ADE detection may not fully capture their incidence and impact.
- Electronic medical records (EMRs) offer potential for automated ADE identification.
Purpose of the Study:
- To assess the efficacy of a computer program in identifying ADEs within an ambulatory care environment.
- To compare the effectiveness of four distinct computer search strategies: diagnosis codes, allergy rules, event monitoring rules, and text searching.
Main Methods:
- A retrospective analysis was conducted on one year of EMR data from 15,665 patients receiving care.
- The study evaluated the presence of ADEs and the sensitivity and specificity of various computer search methods.
Main Results:
- The computer program identified an estimated 864 ADEs, occurring at a rate of 5.5 per 100 patients.
- Approximately 3.4 ADE-related hospital admissions occurred per 1,000 patients.
- Free-text searches demonstrated particular utility, with overall search sensitivity at 58% and specificity at 88%.
Conclusions:
- Computerized search programs, especially those utilizing free-text analysis, are effective tools for detecting ADEs.
- ADEs are frequent and can lead to hospitalizations, underscoring their clinical significance.
- These detection systems add value to EMRs and can support quality improvement initiatives.