Hypothesis: this project will test the hypothesis that automatic external defibrillators (AEDs) used by emergency medical technicians (EMTs) to treat out-of-hospital cardiac arrest, can achieve outcomes (rhythm conversion, admission to hospital, and overall survival rates) comparable to manual defibrillators used by EMTs. Rationale: the most important intervention to resuscitate some of the more than 400,000 people who die annually from sudden cardiac death is early defibrillation. In Emergency Medical Systems where EMTs provide only basic life support, or where paramedic response times are long, a program of early defibrillation by EMTs, produces a large improvement in survival rates. Manual defibrillation by EMTs, however, requires expensive and time-consuming training, special enabling legislation in most states, and acceptance by medical authorities of advanced medical decisions and actions by non-professionals. The full potential of this approach has not been realized. Recently developed AEDs may overcome many barriers to the wide diffusion of defibrillation by EMTs: training and supervision is easier, less expensive, less time-consuming; increased acceptance by medical and EMS authorities. It must first be proven, however, that AEDs, when used by EMTs perform as well as manual defibrillators. Design: a total of 36 EMT units in King County, Washington will participate in a crossover, prospective, randomized, controlled study. 18 units will use AEDs to assess the rhythm in cardiac arrest patients, and to deliver electric countershocks to patients in ventricular fibrillation. 18 other units will use the standard manual defibrillators that require EMT-identification of the rhythm and an EMT-decision to deliver a countershock. After 75 days the 36 units will crossover to the other type of defibrillator. This design eliminates operator variables as each device will be used by the same operators, and demographic variables, as each device will be used in the same geographic locations. Seven of these 75-day cycles will occur. The allocation of patients to treatment by one device or the other will occur in a random fashion consistent with the random pattern of sudden cardiac death in King County.
