Emergency medical services agency-level variation in non-initiation of resuscitation and termination of resuscitation in out-of-hospital cardiac arrest
In brief
EMS agencies showed 4.1-fold variation in odds of ending resuscitation
In 560,240 out-of-hospital cardiac arrests, agencies differed substantially in decisions to stop resuscitation: among patients eligible under the Universal Termination of Resuscitation rule, the median odds were 4.1 times higher when similar cases were compared across randomly selected agencies. Odds of not starting resuscitation varied 2.3-fold; the observational study cannot explain why, highlighting the need to identify and reduce unwarranted differences in care.
- Journal
- Resuscitation (Q1)
- Published
- 25 September 2026
- Study design
- Retrospective cohort
- Evidence level
- Level 3, Low (CEBM 3b)
- Authors
- Sriram Ramgopal, Michelle Nassal, Christian Martin-Gill, Remle P Crowe, Ali Treichel, Clifton W Callaway, et al.
- PMID
- 42790860
- DOI
- 10.1016/j.resuscitation.2026.111333
Why clinicians should know about it
- Picked for Emergency Medicine (paper of the day, 1 October 2026): EMS agency variation in non‑initiation and termination of resuscitation
Abstract
BACKGROUND: Variations in the decisions to initiate and terminate resuscitation for out-of-hospital cardiac arrest (OHCA) likely differ across emergency medical services (EMS) systems. We evaluated EMS-agency variation in non-initiation of resuscitation for EMS-assessed OHCA and applying termination of resuscitation (TOR) for those meeting the Universal TOR rule. METHODS: We conducted a retrospective cohort study of patients with non-traumatic OHCA using the 2018-2024 ESO Data Collaborative. Our outcomes were non-initiation of resuscitation and TOR. For each outcome, we quantified the extent of agency-level variation using the median odds ratio (MOR), which represents the median increase in odds when comparing similar encounters treated by two randomly selected EMS agencies, calculated from mixed-effects models that incorporated demographic, arrest, scene, agency, and EMS care variables. RESULTS: Among 560,240 encounters, resuscitation was not initiated in 76,859 (13.7%). Of the 309,048 encounters eligible for the Universal TOR rule, 175,826 (56.9%) underwent TOR. After adjustment for encounter-level factors, the median agency-specific rates were 13.6% (IQR, 11.0%-16.7%) for non-initiation and 54.3% (IQR, 35.5%-67.4%) for TOR. The MOR for non-initiation of resuscitation was 2.30 (95% CI 2.21-2.38), and the MOR for TOR was 4.13 (95% CI 3.87-4.26). CONCLUSION: Even after adjustment for measured patient and agency characteristics, we observed substantial between-agency variation in non-initiation of resuscitation and TOR. Findings highlight the need for further evaluation of prehospital decision-making processes and opportunities to identify and reduce unwarranted variation and improve consistency, equity, and quality of prehospital care.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.