Fentanyl Overdoses Seen in U.S. Emergency Departments, 2020–2024: Who and How Many

Overdose deaths involving synthetic opioids such as fentanyl rose steeply over the past decade, and deaths involving illegally made fentanyl began to fall from mid-2023. Much less is known about nonfatal fentanyl overdoses — yet a person who survives an overdose is more likely to die of one later, so knowing who is affected can guide prevention and recovery support.

CDC researchers used the National Syndromic Surveillance Program (NSSP), which gathers near-real-time emergency department (ED) records — often within 24–48 hours — from about 80% of U.S. EDs, to track suspected nonfatal overdoses involving fentanyl or its analogs, of unintentional or undetermined intent, from October 2020 to March 2024.

Method

Visits were identified from free-text chief complaints and discharge diagnosis codes; they are called “suspected” because codes in syndromic data can be preliminary or missing. The analysis used the 3,056 EDs (62% of 4,969) with consistent, complete data, and measured visits for fentanyl-involved nonfatal overdose per 10,000 ED visits for any cause, quarter by quarter.

The overall trend

Over the period, 86,404 such visits were recorded.

• The rate rose from 1.4 per 10,000 ED visits in October–December 2020 to a peak of 3.5 in July–September 2023 — an increase of 8.7% per quarter.
• It then fell 11.0% per quarter, to 2.9 in January–March 2024.

That peak and decline match the pattern in overdose deaths involving illegally made fentanyl, which also peaked in mid-2023. Both declines are encouraging, but more time is needed to know whether they will last, and why they happened.

Rates by quarter: A, overall; B, age group; C, sex; D, race and ethnicity. Image from CDC’s page.

Who was affected

Age. Rates were highest among adults 25–34 and lowest among those 55 and older. Among adults 25 and older, rates rose from late 2020 to mid-2023 by 9.3%–11.7% a quarter. Among people 15–24 they held steady through 2021, then rose 9.8% a quarter to mid-2023. Rates then fell significantly among people aged 15–24, 25–34 and 55 and older.

Sex. Among males the rate rose from 2.1 to 4.4, peaking at 5.4 in mid-2023; among females, from 0.7 to 1.6, peaking at 1.9. Both rose at similar speeds until mid-2023; the later decline was significant among males but not females.

Race and ethnicity. American Indian and Alaska Native people had both the highest rates — 11.9 per 10,000 ED visits in mid-2023 — and the fastest rise, averaging 9.0% a quarter. Every group except the combined “other race” group rose significantly until mid-2023. Afterward, rates fell significantly among Black, White and other-race patients, but not among American Indian and Alaska Native or Hispanic patients.

These patterns echo other data: opioid-involved nonfatal ambulance encounters are highest among men and adults 25–34, and in 2021 and 2022 overall overdose death rates were highest among males and among American Indian and Alaska Native people, whose rate rose 15.0% between those years.

Emergency departments as a turning point

For some people, the ED is their main contact with health care, which makes it a place to prevent the next overdose:

• Start treatment. Beginning buprenorphine or another medication for opioid use disorder in the ED can quickly stabilize withdrawal and connect patients to ongoing care.
• Hand out naloxone. Naloxone, which reverses opioid overdoses and can be used at home, can go to patients who have just overdosed and to their families. Because more people appear to be refusing ambulance transport, equipping first responders with naloxone and ways to link people to care matters too.
• Treat the whole person. Screening for and treating co-occurring mental health conditions fits the federal Substance Abuse and Mental Health Services Administration’s “No Wrong Door” approach, and peer recovery specialists with lived experience can help people into care.

The need is large: in 2023, in data from 38 jurisdictions, about two-thirds (65.9%) of fatal overdoses from any drug had at least one missed opportunity to intervene — a potential bystander present (42.6%), a mental health diagnosis (28.7%) or an earlier overdose (13.5%) — while fewer than a quarter (23.7%) had any record of naloxone being given.

Limitations

• Fentanyl overdoses may be missed or misclassified because of hospital drug-testing practices.
• The definition cannot tell illegally made fentanyl from prescription fentanyl, though most fentanyl overdose deaths involve the illegal kind.
• Data quality, including demographic data, varies between facilities.
• Only overdoses treated in EDs are counted.
• About 1% of the visits ended in death, so a few fatal overdoses are included.
• Rates among American Indian and Alaska Native people may be underestimated because of racial misclassification, and most tribal health facilities do not report to NSSP.

Sources

Based on Pickens CM, Park J, Casillas SM, et al., "Trends in Suspected Fentanyl-Involved Nonfatal Overdose Emergency Department Visits, by Age Group, Sex, and Race and Ethnicity — United States, October 2020–March 2024," Morbidity and Mortality Weekly Report, Centers for Disease Control and Prevention; a work of the United States government in the public domain.