Introduction: Previous studies evaluating whether subsequent conversion to shockable rhythms in patients who had initially non-shockable rhythms was associated with altered clinical outcome reported inconsistent results. Therefore, we hypothesized that subsequent shock delivery by emergency medical service (EMS) providers altered clinical outcomes in patients with initially non-shockable rhythms. Methods: We tested for an association between subsequent shock delivery in EMS resuscitation and clinical outcomes in patients with initially non-shockable rhythms (n = 11,481) through a survey of patients after out-of-hospital cardiac arrest in the Kanto region (SOS-KANTO) 2012 study cohort, Japan. The primary investigated outcome was 1-month survival with favorable neurological functions. The secondary outcome variable was the presence of subsequent shock delivery. We further evaluated the association of interval from initiation of cardiopulmonary resuscitation to shock with clinical outcomes. Results: In the univariate analysis of initially non-shockable rhythms, patients who received subsequent shock delivery had significantly increased frequency of return of spontaneous circulation, 24-hour survival, 1-month survival, and favorable neurological outcomes compared to the subsequent not shocked group (P &lt
0.0001). In the multivariate logistic regression analysis, subsequent shock was significantly associated with favorable neurological outcomes (vs. not shocked
adjusted P = 0.0020, odds ratio, 2.78
95 % confidence interval, 1.45-5.30). Younger age, witnessed arrest, initial pulseless electrical activity rhythms, and cardiac etiology were significantly associated with the presence of subsequent shock in patients with initially non-shockable rhythms. Conclusions: In this study of cardiac arrest patients with initially non-shockable rhythms, patients who received early defibrillation by EMS providers had increased 1-month favorable neurological outcomes.