Opioid misuse continues to pose a critical public health issue in the United States, particularly due to the increasing prevalence of synthetic opioids such as fentanyl.1 Efforts to curb this epidemic have been implemented, such as increased fentanyl screening and increasing access to naloxone2. However, questions remain about whether fentanyl testing during ED encounters is associated with subsequent treatment and survival.
To understand trends in overdose visits, we studied more than 230 million ED visits between January 2018 and November 2025. We studied overall overdoses, overdoses from opioids, and overdoses from synthetic opioids, each a subset of the preceding broader category.
The 12-month rolling average rate of ED visits for all types of overdoses peaked at 946 per 100,000 in March 2021, a period when overall ED volumes were unusually low, as seen in Figure 1.3 By November 2025, the rate had declined to 634 per 100,000, reflecting a 33% decline and falling below pre-pandemic levels. Opioid overdoses followed a similar trajectory, with rates trending downward after naloxone became available over the counter in 2023.2 In contrast, synthetic opioid overdoses increased more than threefold from 15 to 51 per 100,000 visits between December 2018 and November 2025, though rates have declined after peaking at 66 per 100,000 in December 2023. Notably, testing for synthetic opioids also increased by more than 400% from the first quarter of 2021 to the fourth quarter of 2025, as seen on the Epic Research Fentanyl and Opiate Toxicology data tracker.4
There has been a rise in the incidence of synthetic opioid overdose mortality cases since 2018,5 picking up in 2020 and reaching a peak of 77,695 cases in June 2023, according to the U.S. Centers for Disease Control and Prevention (CDC), as seen in Figure 2. Since then, the number of synthetic opioid overdose deaths has declined substantially, reaching a low of 38,514 cases in August 2025, a reduction of more than 50% from the peak.
To understand the relationship between fentanyl testing and outcomes such as mortality and repeat overdose encounters, we studied more than 1 million patients who had a non-fatal ED visit for an overdose. We accounted for demographics, social vulnerability and rurality based on most recently documented address, count of prior overdose encounters, administration of naloxone during the ED visit, and ED acuity score.
Among patients who were tested for fentanyl, the likelihood of being prescribed a medication for opioid use disorder (MOUD) during the encounter was 147% higher compared to those who were not tested for fentanyl, as seen in Figure 3. Short term outcomes were also improved: the likelihood of death within 30 days was 11% lower. There was not a significant difference in the likelihood of a repeat ED visit for an overdose within 30 days between patients who had fentanyl testing and those who did not. The likelihood of mortality or a repeat overdose within one year was higher among those who had fentanyl testing than those who were not tested. This pattern might reflect that patients who receive fentanyl testing represent a higher-risk population with more severe substance use disorders.
Importantly, fentanyl testing might indicate clinical suspicion or resource availability. Indications for testing patients for fentanyl might include additional considerations we were not able to fully assess as part of this analysis.
