Abstract Body: BACKGROUND: Bystander cardiopulmonary resuscitation (CPR) and defibrillation improves survival in out-of-hospital cardiac arrest. However, much of the community do not have recent training in these basic life support (BLS) skills, particularly culturally diverse populations.

AIM: The FirstCPR cluster randomised controlled trial aimed to test the effectiveness of a community organisation-based BLS education and training approach delivered both digitally and in-person.

METHODS: FirstCPR was conducted in New South Wales, Australia, between May 2021 and December 2023. Clusters (community organisations with 50+ members), stratified by type, size, and location were randomly allocated to intervention (intervention for 12 months) or control (delayed after 12 months). Analysis (intention-to-treat) was performed using a hierarchical mixed models approach with random effects for the cluster and fixed effects for organisation and participant characteristics. The primary outcome was 'training and willingness-to-perform CPR' on a stranger - assessed in surveys of participating community organisation members at 12 months.

RESULTS: Of 165 clusters (82 randomised to intervention), 58% were sports groups and 42% were social or faith-based groups (>50% culturally diverse). On survey, among intervention organisation members at study initiation, 12% reported recent training (i.e., in the last 12 months). Intervention implementation was impacted by the COVID-19 pandemic and less face-to-face training was delivered and a more detailed electronic campaign was implemented. Evaluation surveys were completed by 407 members from intervention organisations and 517 members in control organisations. FirstCPR intervention doubled the likelihood of participants from intervention organisations reporting they were trained and willing to perform CPR on a stranger (primary outcome) (62.7% vs 44.9 %, p<0.001; adjusted Relative Risk 2.22: 95%CI: 1.50 – 3.30) (Figure 1). Confidence in using an automated external defibrillator and willingness to use it on a stranger was also significantly higher in intervention clusters.

CONCULSIONS: Our novel approach, using a broad range of established community organisations, could be an effective population-wide approach to improving BLS education to enable reach into diverse, underserved groups.