Background:Auto‑verification enables medication orders to bypass pharmacist review and verify automatically upon entry. Its use is limited to settings, such as the emergency department (ED), where a licensed practitioner oversees ordering, preparation, and administration. While this workflow improves efficiency in emergency departments (EDs), there are concerns regarding the safety of auto‑verifying high‑alert medications, which when used incorrectly can cause significant patient harm. At Cone Health, certain high‑alert agents, including opioids such as oxycodone, are permitted to be auto‑verified. Opioids carry risks for respiratory depression, overdose, and dizziness/lightheadedness leading to instability. A quality improvement evaluation was performed to assess whether auto‑verification of oxycodone-containing medications results in clinically meaningful errors or overlooked interventions.
Methods:This IRB reviewed, determined exempt retrospective chart review included auto‑verified oxycodone immediate‑release 5 mg tablets and oxycodone/acetaminophen 5/325 mg tablets ordered in Cone Health EDs between January 1 and July 31, 2025. There were no exclusion criteria. A random sample of 140 orders (70 from each product) was selected using random date and order number generation. Outcomes measured in this study were incidence of order modification, prevalence of adverse effects, and quantity of documented pharmacist interventions. Adverse effects evaluated included overdose (naloxone use as a proxy) and allergic reaction.
Results:A total of 9,456 auto‑verified orders were identified during the study period. Among the 140 orders analyzed, 82% (n = 140) had at least one modification during
initial order entry. Frequency was the most modified order parameter (79%, n= 140). Dispense location changes occurred in a further 12% (n = 140) of all orders. No dose, route, start time, or stop time adjustments were observed. Oxycodone/acetaminophen orders had a higher rate of modification compared with oxycodone IR (91% vs. 73%). None of the order modifications were made by a pharmacist. Rather, order modifications were made by providers and mid-level providers. Additionally, there were no documented pharmacist interventions and no adverse effects in the sample.
Conclusion:Auto‑verification of oxycodone products in the Cone Health ED was not associated with patient harm. Although modifications at initial order entry were common, these changes did not lead to errors. Expansion of this quality improvement process to additional opioids and other high‑alert medications is warranted.
