Introduction : In 2005, the American Heart Association recommended increased “hands-on time” during cardiopulmonary resuscitation (CPR) based on animal research and small case series. This study estimated the effect of increasing CPR fraction (proportion of resuscitation time with active CPR) on survival in a cohort of patients with out-of hospital ventricular fibrillation or ventricular tachycardia (VF/VT). Methods: Patients were selected from the ROC Epistry who had a confirmed VF/VT cardiac arrest that was not witnessed by Emergency Medical Services (EMS), received no public access defibrillation shock prior to EMS arrival, and had impedance recordings of CPR before the first shock. The proportion of each minute with active CPR, from defibrillator pad application until the first shock, was measured from the electronic resuscitation record by skilled readers who were blinded to hospital discharge outcome. The effect of increasing CPR fraction on survival to hospital discharge was adjusted for age, sex, bystander CPR, public location, interval from 911 call to defibrillator activation, chest compression rate, and ROC community. Results: Of 7963 EMS-treated cases of cardiac arrest without public access defibrillation, 1893 had an initial rhythm of VF/VT and 283 of those had electronic tracings and confirmed outcome. Mean age was 63 years and 81% were male. Bystanders performed CPR on 51% and 41% arrested in a public location. Outcomes and odds ratios (OR) with 95% confidence intervals (CI) of survival are shown from lowest to highest category of CPR fraction. Conclusions: This study provides preliminary evidence that increasing CPR fraction is associated with increased survival from VF/VT cardiac arrest. Though the study was observational, relatively small, and able only to measure CPR fraction after pad application, these findings suggest that provision of minimally interrupted CPR has direct clinical impact on survival after VF cardiac arrest.