Treatment of cardiac arrest with rapid defibrillation by police in King County, Washington.

BACKGROUND: Improving survival from out-of-hospital cardiac arrest is an ongoing challenge for emergency medical services (EMS). Various strategies for shortening the time from collapse to defibrillation have been used, and one is to equip police officers with defibrillators. Objective. We evaluated the programmatic implementation of police defibrillation to determine if such a program could improve the process of care in a high-functioning and mature EMS system.

METHODS: We conducted a prospective observational study of implementation of a police defibrillation in two police departments in King County, Washington, from March 1, 2010 to March 31, 2012. The program was designed to dispatch police specifically to cases with a high suspicion of cardiac arrest, defined as a patient who was unconscious and not breathing normally. We included all nontraumatic out-of-hospital cardiac arrest events that occurred prior to EMS arrival and within the city limits of the two cities. We collected both EMS and police dispatch reports to document times of call receipt, dispatch, and arrival of both agencies. We obtained rhythm recordings when the automated external defibrillators (AEDs) were used by the police. Descriptive statistics were used to measure frequency of police dispatch and to compare times to treatment between patients with a police response and those without.

RESULTS: During the study period there were 231 cases of cardiac arrest that occurred prior to EMS arrival eligible for police response in the study communities. Police were dispatched to 124 (54%) of these cases. Of the 124, the police arrived before EMS 37 times, or 16% of the 231 cases. Police performed CPR in 29 of these cases and applied the AED in 21 of them. Of the 21 cases in which the AED was applied for cardiac arrest, a shock was delivered on first analysis for 6 patients. Although the response interval between dispatch to scene arrival was similar for EMS and police (4.5 minutes versus 4.6 minutes respectively, p = 0.08), police were dispatched considerably slower than EMS (1.8 minutes versus 0.6 minutes, p < 0.001).

CONCLUSIONS: In the current programmatic implementation, police had a measurable but limited involvement in resuscitation. Efforts to address dispatch challenges may improve police involvement.