2025 Southeastern Residency Conference

TITLE: Auto-verification Resulting in Medication Errors in the Emergency Department


AUTHORS: Courtney Ellison, Nichole Moore, Rachel Rossi, Christopher Whitman, Rachel Foster, Amanda Williams, and Maria Berec


BACKGROUND: Auto-verification is a process commonly utilized by hospitals to increase workflow. The Joint Commission and the American Society of Health-System Pharmacists (ASHP) provide standards to incorporate into auto-verification software, such as “do not verify” criteria, to ensure safety. Available literature has demonstrated the proficiency of auto-verification; however, limited studies have evaluated its safety. This study aimed to assess current medications auto-verified in a health system’s emergency department for the rate of errors and compare them to medications verified by a pharmacist while observing the differences in time of verification and time of administration of first doses.

METHODS: A multi-site retrospective cohort study of auto-verified adult and pediatric medication orders from July 2023 through September 2023 was conducted. Data for antibiotic and anticoagulant medication orders auto-verified in patients ≥ 19 years old and auto-verified pediatric (≤ 18 years old) orders were collected and compared to orders verified by pharmacists. Orders entered by pharmacists were excluded from the study. Medication order numbers were placed in a random list generator to identify comparator groups and evaluated based on medication, dose, frequency, indication, and if duplicates were present. Other data collected included patient demographics, location, time of ordering, time to verification, and time to administration. The primary endpoint was the rate of medication errors, which were classified based on the type of error. Secondary endpoints included near misses, average ordering, administration, and verification time. Fisher's exact test and relative risk ratios used to analyze the primary endpoint.

RESULTS: A total of 1,003 medication orders were retrospectively evaluated between July 2023 through September 2023. Comparator groups consisted of 251 samples per group: adult auto-verified antibiotics and anticoagulants, adult pharmacy-verified antibiotics and anticoagulants, and pediatric auto-verified medications, with 250 samples in the pediatric pharmacy-verified medications group. Thirty-two medication errors were identified among the adult auto-verified group, and 10 in the adult pharmacy-verified group. The rate of medication errors among the pediatric orders auto-verified and pharmacy-verified orders were 14 and 4, respectively. Incorrect dosing accounted for most errors identified among the auto-verified adult (50%; 16/32) and pediatric orders (79%; 11/14). For the primary endpoint there was a relative risk of 3.3 (95% CI 1.66 – 6.55; p=0.0003) for the adult population and a relative risk of 3.48 (95% CI 1.16 – 10.44; p=0.028) for the pediatric population. The average times to order verification were 13.2 minutes for adults and 6.9 minutes for pediatrics. The average times to administration for auto-verified orders were 29.1 minutes in adults and 18.6 minutes in pediatrics. Of all the medication errors identified, 21.7% (10/46) were considered near misses.

CONCLUSION: Auto-verification can provide proficient patient care in fast-paced settings where accuracy is vital. Periodic evaluation of this process is essential to evaluate the safety of present standards. Among medications auto-verified in the emergency department, there was a statistically significant difference compared to pharmacy-verified medications in adult and pediatric patients. Limitations identified were a small percentage of patients’ renal functions were evaluated among auto-verified orders (49% adults and 18% pediatrics), and most orders were one-time doses. The randomization process was considered a limitation because medications in the auto-verified groups were inconsistent with pharmacy-verified groups in terms of specific drugs. The results of this study will be utilized for quality improvement purposes for auto-verification criteria and safety evaluation of other drug classes commonly auto-verified.