Differential association of prehospital physician presence with outcomes in out-of-hospital cardiac arrest: Secondary analysis of a nationwide registry

OBJECTIVES: Although prehospital physician involvement may affect outcomes after out-of-hospital cardiac arrest (OHCA), its effect could vary across patient subgroups. We aimed to evaluate whether physician presence is associated with improved survival in subgroups of patients with OHCA. METHODS: We used an observational study design and a retrospective secondary analysis of data from the nationwide Japanese Association for Acute Medicine OHCA registry (2014-2022). Adult patients (age ≥18 years) resuscitated after hospital arrival with complete prehospital physician data were included. The primary outcome was favorable neurological status at 30 days, defined as a cerebral performance category score of 1 or 2. Multivariate logistic regression was performed to assess the association between physician presence and outcomes, adjusting for clinically relevant confounders. Prespecified subgroup analyses were also performed according to age, witness status, bystander intervention, etiology, and extracorporeal cardiopulmonary resuscitation (ECPR) use. RESULTS: Among 83,239 patients, 9,562 had physician attendance. Physician presence was associated with favorable neurological outcomes at 30 days (adjusted odds ratio 1.49, 95% confidence interval 1.34-1.65). The association was consistent across most prespecified subgroups and was stronger among patients with unwitnessed arrest; however, no significant association was observed in patients who underwent ECPR. CONCLUSIONS: Prehospital physician presence was associated with improved neurological and survival outcomes after OHCA, with variation across patient subgroups; however, given the observational design and potential selection bias, these findings should be regarded as hypothesis-generating. Further prospective studies are warranted to define patient subgroups that benefit most from physician presence on scene and to optimize dispatch protocols accordingly.