mepa8 Get alerts
EMS/Breaking

RACE-CARS trial in 60 North Carolina counties did not improve cardiac arrest survival

An exploratory analysis found 33% higher odds of bystander CPR in intervention counties during the trial's final year, the AHA release said.

Mepa8 headline card: RACE-CARS trial in 60 North Carolina counties did not improve cardiac arrest survival

Key takeaways

  • The RACE-CARS cluster-randomized trial in North Carolina found survival with good neurologic outcome of 9.0% in intervention counties versus 9.4% with usual care, a difference that was not statistically significant [1].
  • The intervention bundled 911 telecommunicator training for arrest recognition and CPR instructions, community CPR classes, and expanded AED use by first responders and bystanders before EMS arrival [1].
  • Bystander CPR was 44.8% in intervention counties and 42.3% in usual-care counties; defibrillation before EMS arrival was 13.5% versus 15.3% [2].
  • An exploratory analysis found 33% higher odds of bystander CPR in intervention counties during the final study year, which the AHA release called a reason for further study [2].
  • For crews: the trial provided 405 AEDs to 31 counties, and 18 of those counties deployed them with fire departments and law-enforcement agencies [2].

A cluster-randomized trial that trained 911 telecommunicators, taught CPR across communities and added automated external defibrillators in North Carolina counties did not improve survival after out-of-hospital cardiac arrest, according to a paper in The New England Journal of Medicine published alongside a November 8, 2026 presentation at the American Heart Association's Scientific Sessions in Chicago [1][2]. The trial is called RACE-CARS and was funded by the National Heart, Lung, and Blood Institute [1].

The abstract reports that 9.0% of patients in intervention counties survived to hospital discharge with a good neurologic outcome, compared with 9.4% in counties that continued usual care [1]. The adjusted odds ratio was 0.97, with a 95% confidence interval of 0.75 to 1.26 and a P value of 0.80 [1].

Christopher Granger, the trial's principal investigator and a cardiologist and professor at Duke University, said in the AHA release that a late rise in bystander CPR "suggests that a sustained community outreach may be needed to show an improvement in survival" [2].

What the records show

The trial was registry-based and drew its data from the Cardiac Arrest Registry to Enhance Survival [1]. It ran from July 1, 2022, to June 30, 2026 [1]. North Carolina counties were grouped into clusters and randomized 1:1 to the intervention or to usual care [1]. The AHA release puts the count at 60 counties with a combined population of over 8.5 million people [2].

The intervention had four components, per the NEJM abstract [1]:

  • Training 911 telecommunicators to recognize cardiac arrest faster.
  • Training telecommunicators to deliver better CPR instructions to callers.
  • Community CPR training.
  • Expanded use of automated external defibrillators by first responders or bystanders before EMS arrived.

The primary outcome was survival to hospital discharge with a Cerebral Performance Category score of 1 or 2 on a scale of 1 to 5 [1]. Patients had a median age of 64.0 years, and 36.7% were women [1].

MeasureInterventionUsual care
Clusters2827
Patients11,61510,844
Survival with good neurologic outcome9.0% (1048 of 11,608)9.4% (1017 of 10,839)
Bystander CPR44.8%42.3%
Defibrillation before EMS arrival13.5%15.3%

Sources: NEJM abstract [1] for clusters, patients and the primary outcome; AHA release [2] for bystander CPR and pre-arrival defibrillation.

The AHA release also describes how the program was rolled out. The trial supplied 405 defibrillators to 31 counties [2]. Of those counties, 26 (87%) deployed the devices, 19 (63%) placed them in publicly accessible locations, and 18 (60%) put them with first responders such as fire departments and law-enforcement agencies [2].

Community partners held 5,724 CPR training events that reached 201,586 people, or 4.8% of residents in the intervention communities [2]. Reach varied widely. Four counties trained more than 10% of residents, and eight trained fewer than 3% [2]. Training was free and was built into schools, workplaces and county government, the release said [2].

In an exploratory analysis, people who arrested in intervention communities during the final study year had 33% higher odds of receiving bystander CPR than those in usual-care communities [2]. The release said this finding "warrant[s] further study" and labeled it exploratory [2].

Why it matters for EMS

The trial tested the interval before an ambulance arrives, which is the part of the chain that dispatch centers, first responders and public AED programs are built to shorten. Study author Carolina Malta Hansen, a cardiologist at Herlev and Gentofte Hospital and Copenhagen EMS in Denmark, said the trial reached "rural areas where emergency care often relies on volunteer-first responders," according to the release [2]. Hansen also said in the release that the chance of survival "decreases by 10% every minute without CPR" [2].

Defibrillation before EMS arrival was lower in intervention counties than in usual-care counties despite the added devices, at 13.5% versus 15.3% [2]. The release states that the program did not improve overall rates of bystander CPR or defibrillation [2]. EMS clinicians follow local protocols and medical direction.

Context

The AHA release cites the association's 2026 Heart Disease and Stroke Statistics, which put out-of-hospital cardiac arrests in the U.S. at approximately 350,000 per year with about 1 in 10 surviving [2]. It also cites the association's AED Fact Sheet, which says only about 40% of these patients receive bystander CPR and only 10% have an AED applied by a bystander before EMS arrives [2]. The association's 2025 Guideline for CPR and emergency cardiovascular care emphasizes early recognition, calling 9-1-1, CPR and AED use, the release said [2].

What the records do not tell us

The packet contains the NEJM abstract, not the full paper [1]. The abstract does not report secondary outcomes, and the release gives bystander CPR and defibrillation rates without confidence intervals or significance tests [1][2]. The 33% figure for the final year comes from an exploratory analysis, and the release does not give its confidence interval [2].

Neither record reports how fully the telecommunicator training was delivered, how often dispatchers recognized arrest, or how often callers received CPR instructions [1][2]. The records do not explain why pre-arrival defibrillation was lower in intervention counties [2]. The abstract's primary-outcome denominators of 11,608 and 10,839 are smaller than the enrolled counts of 11,615 and 10,844, and the abstract does not say why [1]. The release describes the presented findings as preliminary and says it includes updated information from the researchers that is not in the abstract [2]. Overall survival regardless of neurologic status is not reported in either record [1][2].

What to watch

Granger said in the release that the investigators "have established an efficient, high-quality clinical trial network to continue this work" [2]. Hansen said, "Changing how an entire community responds to cardiac arrest takes time, and our findings show just how challenging that can be" [2]. The records do not name a follow-on trial or a date for further results [1][2].

Questions readers ask

What was the RACE-CARS trial?

RACE-CARS was a registry-based, cluster-randomized trial in North Carolina counties that tested a four-part intervention: 911 telecommunicator training for arrest recognition, telecommunicator CPR instruction training, community CPR training, and expanded AED use before EMS arrival. It ran from July 1, 2022, to June 30, 2026, and was funded by the National Heart, Lung, and Blood Institute.

Did the RACE-CARS intervention improve cardiac arrest survival?

No. Survival to hospital discharge with good neurologic outcome was 9.0% in intervention counties and 9.4% in usual-care counties, with an adjusted odds ratio of 0.97 and a P value of 0.80, according to the NEJM abstract.

Did RACE-CARS increase bystander CPR?

Overall bystander CPR was 44.8% in intervention counties versus 42.3% in usual-care counties, which the AHA release said was not a significant improvement. An exploratory analysis found 33% higher odds of bystander CPR in intervention counties during the final study year.

How many patients were in the RACE-CARS trial?

The trial included 11,615 patients in 28 intervention clusters and 10,844 patients in 27 usual-care clusters, drawn from 60 North Carolina counties with a combined population of over 8.5 million people.

Sources

  1. NEJM abstract via PubMed: Community and EMS Interventions to Improve Survival in Cardiac Arrest — The New England journal of medicine (2026-11-09). pubmed.ncbi.nlm.nih.gov/43001758/
  2. American Heart Association news release, Nov. 8, 2026 — newsroom.heart.org (2026-11-08). newsroom.heart.org/news/community-cpr-and-ems-intervention-program-did
How this story was made Drafted by Mepa8's AI newsroom from the 2 records listed above and nothing else. Before publication a verification program confirmed that every number and quotation in the story appears in those records, that no passage was copied from them, and that every cited link opens. Editor: Ryan Bolden, EMS professional, who reviews published stories and handles corrections. Medical review pending. How we use AI · Report an error

This is news, not medical advice. EMS clinicians: follow your local protocols and medical direction.

Free alerts

Know about the next recall or outbreak the day it is posted.

One email when a story touches EMS equipment, medications or public health. Plus the free 6am briefing if you want it.

Send me