A multifaceted intervention to increase bystander CPR rates after out-of-hospital cardiac arrest in France: a stepped-wedge, cluster-randomised, controlled trial
Abstract
BACKGROUND: Bystander cardiopulmonary resuscitation (CPR) is a key determinant of survival after out-of-hospital cardiac arrest (OHCA), but rates remain suboptimal worldwide. We evaluated whether a multifaceted intervention targeting emergency dispatch systems and community response could increase bystander-initiated CPR before emergency medical service (EMS) arrival. METHODS: We conducted a stepped-wedge, cluster-randomised trial in 15 French EMS agencies. The intervention combined dispatcher training for early recognition of cardiac arrest and telephone-assisted CPR, activation of trained volunteers through a geolocation smartphone application, and structured audit and feedback provided to EMS agencies. Consecutive adults (≥18 years) with non-traumatic OHCA were included. Randomisation was stratified by tercile of OHCA volume managed by each EMS in 2015. The intervention was implemented during a 1-month transition period before the intervention period at each EMS centre. A dedicated protocol guided dispatchers to identify cardiac arrest from unresponsiveness and abnormal breathing, then gave scripted compression-only instructions and activated volunteers, targeting recognition within 60 s of call pick-up and first compression within a further 60 s. The primary outcome was bystander-initiated CPR before EMS arrival. The trial was prospectively registered on ClinicalTrials.gov (NCT03633370) and is completed. FINDINGS: Of the 15 EMS agencies initially randomised, two withdrew before recruitment began and were replaced, and one enrolled no patients. Therefore, between Aug 27, 2018, and Feb 20, 2022, 14 EMS agencies recruited a total of 2160 patients. The intention-to-treat analysis included 1684 patients (686 control, 998 intervention) enrolled across the 12 originally randomised EMS agencies. The mean age for all patients included in the intention-to-treat analysis was 67 years (SD 15), and 1205 (71·6%) were male. Bystander-initiated CPR before EMS arrival occurred in 421 (61·4%) patients in the control group and 708 (70·9%) patients in the intervention group (adjusted odds ratio [OR] 1·54 [95% CI 1·09-2·19]; p=0·017). Estimates were consistent across six prespecified sensitivity analyses (OR range, 1·38-1·55), remaining significant in four, but not after adjustment for cluster-specific secular trends (1·38 [0·98-1·94]) or baseline characteristics (1·39 [0·97-1·98]). No intervention-related adverse events were reported. INTERPRETATION: A multifaceted intervention was associated with increased bystander-initiated CPR after OHCA, although the estimate was sensitive to the modelling of secular trends, and it did not translate into improved clinical outcomes. These findings support further investigation of coordinated community-based strategies to strengthen the early links in the chain of survival at a population level. FUNDING: Direction Générale de l'Offre de Soins.
Related
- Physiotherapy Educational Intervention for Occipital Positional Plagiocephaly Prevention: a Multicentre Randomised Controlled Trial
- Community-tailored One Health educational intervention to enhance knowledge and practices for zoonotic disease prevention in rural Thailand: A protocol for a prospective cluster randomised controlled Trial in Chanthaburi, Thailand (Saan Suk trial)
- Which changes in adolescents' knowledge, attitudes, and practices following a school-based intervention in France increase HPV vaccination uptake and intentions? A mediation analysis
- Effect of different forms of upper limb muscle training on dyspnea in chronic obstructive pulmonary disease: a randomised controlled trial
- Efficacy and tolerability of a cocktail of bacteriophages to treat burn wounds infected by Pseudomonas aeruginosa (PhagoBurn): a randomised, controlled, double-blind phase 1/2 trial
BibTeX
@article{debaty2026multifaceted,
title = {{A multifaceted intervention to increase bystander CPR rates after out-of-hospital cardiac arrest in France: a stepped-wedge, cluster-randomised, controlled trial}},
author = {Guillaume Debaty and Simon Ribes and Patrick Ray and Tahar Chouihed and Nicolas Desseigne and Pascal Usseglio-Polatera and Patrice Serre and Catherine Pradeau and Vincent Bounes and Thierry Roupioz and Anne-Sophie Lucas and Didier Giolito and Claude Zamour and Christophe Boyer and Hervé Hubert and Dominique Savary and Karim Tazarourte and Yonathan Freund and Samuel I Watson and Karla Hemming and Lionel Lamhaut and José Labarère},
journal = {The Lancet Public Health},
year = {2026},
month = nov,
doi = {10.1016/s2468-2667(26)00200-8},
url = {https://hal.science/hal-05850225v1},
note = {CC BY},
}
RIS
TY - JOUR
TI - A multifaceted intervention to increase bystander CPR rates after out-of-hospital cardiac arrest in France: a stepped-wedge, cluster-randomised, controlled trial
AU - Guillaume Debaty
AU - Simon Ribes
AU - Patrick Ray
AU - Tahar Chouihed
AU - Nicolas Desseigne
AU - Pascal Usseglio-Polatera
AU - Patrice Serre
AU - Catherine Pradeau
AU - Vincent Bounes
AU - Thierry Roupioz
AU - Anne-Sophie Lucas
AU - Didier Giolito
AU - Claude Zamour
AU - Christophe Boyer
AU - Hervé Hubert
AU - Dominique Savary
AU - Karim Tazarourte
AU - Yonathan Freund
AU - Samuel I Watson
AU - Karla Hemming
AU - Lionel Lamhaut
AU - José Labarère
PY - 2026
DA - 2026/11/03
JO - The Lancet Public Health
DO - 10.1016/s2468-2667(26)00200-8
UR - https://hal.science/hal-05850225v1
AB - BACKGROUND: Bystander cardiopulmonary resuscitation (CPR) is a key determinant of survival after out-of-hospital cardiac arrest (OHCA), but rates remain suboptimal worldwide. We evaluated whether a multifaceted intervention targeting emergency dispatch systems and community response could increase bystander-initiated CPR before emergency medical service (EMS) arrival. METHODS: We conducted a stepped-wedge, cluster-randomised trial in 15 French EMS agencies. The intervention combined dispatcher training for early recognition of cardiac arrest and telephone-assisted CPR, activation of trained volunteers through a geolocation smartphone application, and structured audit and feedback provided to EMS agencies. Consecutive adults (≥18 years) with non-traumatic OHCA were included. Randomisation was stratified by tercile of OHCA volume managed by each EMS in 2015. The intervention was implemented during a 1-month transition period before the intervention period at each EMS centre. A dedicated protocol guided dispatchers to identify cardiac arrest from unresponsiveness and abnormal breathing, then gave scripted compression-only instructions and activated volunteers, targeting recognition within 60 s of call pick-up and first compression within a further 60 s. The primary outcome was bystander-initiated CPR before EMS arrival. The trial was prospectively registered on ClinicalTrials.gov (NCT03633370) and is completed. FINDINGS: Of the 15 EMS agencies initially randomised, two withdrew before recruitment began and were replaced, and one enrolled no patients. Therefore, between Aug 27, 2018, and Feb 20, 2022, 14 EMS agencies recruited a total of 2160 patients. The intention-to-treat analysis included 1684 patients (686 control, 998 intervention) enrolled across the 12 originally randomised EMS agencies. The mean age for all patients included in the intention-to-treat analysis was 67 years (SD 15), and 1205 (71·6%) were male. Bystander-initiated CPR before EMS arrival occurred in 421 (61·4%) patients in the control group and 708 (70·9%) patients in the intervention group (adjusted odds ratio [OR] 1·54 [95% CI 1·09-2·19]; p=0·017). Estimates were consistent across six prespecified sensitivity analyses (OR range, 1·38-1·55), remaining significant in four, but not after adjustment for cluster-specific secular trends (1·38 [0·98-1·94]) or baseline characteristics (1·39 [0·97-1·98]). No intervention-related adverse events were reported. INTERPRETATION: A multifaceted intervention was associated with increased bystander-initiated CPR after OHCA, although the estimate was sensitive to the modelling of secular trends, and it did not translate into improved clinical outcomes. These findings support further investigation of coordinated community-based strategies to strengthen the early links in the chain of survival at a population level. FUNDING: Direction Générale de l'Offre de Soins.
SN - 2468-2667
ER -