In a randomized controlled clinical trial, the effectiveness of emergency medical technician (EMT) use of automatic external defibrillators (AEDs) was compared with EMT use of standard defibrillators for patients in cardiac arrest. A total of 321 cardiac arrest patients were treated during the study: 116 were treated by EMTs using the AED (AUTO group), 158 were treated by EMTs using the standard defibrillators (standard group), and 47 were treated by EMTs using the standard defibrillator when they were assigned to use the AED. There was no significant differences in hospital admission or discharge rates between the AUTO group (54% admitted, 28% discharged) and the standard group (52% admitted, 23% discharged) for patients in ventricular fibrillation (VF), for patients in non-VF rhythms, or for all patients combined. The only significant difference observed was in the time from power ON to first shock: 1.1 minutes average AUTO group and 2.0 minutes average standard group. The treatment groups did not differ significantly in sensitivity for VF (78% AED, 76% standard), specificity for non-VF rhythms (100% AED, 95% standard), or rates of defibrillation to a non-VF rhythm (62% AED, 57% standard). We conclude that in clinical outcomes and device performance, AEDs are comparable with standard defibrillators and should be considered an acceptable alternative. Automatic external defibrillators appear to have advantages over standard defibrillators in training, skill retention, and faster operation. Such devices can make early defibrillation available for a much larger portion of the population. They are a major innovation for the prehospital care of cardiac arrest patients.