[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
September 20, 2000

The Shocking Truth About Automated External Defibrillators

Author Affiliations

Author Affiliations: Department of Emergency Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Mass (Dr Brown); and the Department of Emergency Medicine, Emory University School of Medicine, Atlanta, Ga (Dr Kellermann).


Controversies Section Editor: Phil B. Fontanarosa, MD, Executive Deputy Editor.

JAMA. 2000;284(11):1438-1441. doi:10.1001/jama.284.11.1438

The first successful electrical defibrillation was described in 1947, when Claude Beck used open-chest massage and alternating current internal defibrillation to resuscitate a 14-year-old boy whose heart was in ventricular fibrillation.1 Immediate defibrillation is now considered the standard of care for treatment of ventricular fibrillation and should be undertaken before initiation of other advanced life support measures, such as endotracheal intubation. It has been clearly shown that the interval from cardiac arrest to defibrillation is a major determinant of successful resuscitation and survival to hospital discharge. With each minute that defibrillation is delayed, the chances of successful resuscitation decrease by 2% to 10%.2 It is therefore vital that defibrillation be attempted as early as possible.

In response to this observation, extensive efforts have been made to shorten the time from collapse to defibrillation in cases of out-of-hospital cardiac arrest. Much of this activity has focused on placing a defibrillator in the hands of an increasing number of non–hospital-based care providers. Shortly after researchers demonstrated that specially trained nonphysicians could safely and effectively operate a manual defibrillator,3,4 efforts were initiated to determine if out-of-hospital defibrillation could be safely and effectively performed by individuals with less sophisticated levels of training.

In 1980, using a quasi-experimental design, Eisenberg and colleagues5 demonstrated that patients who received care from basic-level emergency medical technicians (EMTs) who were taught to provide out-of-hospital defibrillation achieved significantly higher rates of survival to hospital discharge than patients cared for by EMTs who provided only cardiopulmonary resuscitation (CPR) and rapid transport to a hospital. Four years later, Eisenberg et al6 showed that basic-level EMTs could use a manual defibrillator to deliver countershocks prior to paramedic arrival in a tiered EMS system. It proved difficult, however, for rural ambulance services that rarely perform defibrillation to ensure that their EMTs received the regular training needed to retain acceptable proficiency in manual defibrillation.7

Experience With AEDs

When automated external defibrillators (AEDs) were introduced, EMT defibrillation received a substantial boost. Automated external defibrillators use a microcomputer to analyze the victim's cardiac rhythm and advise the rescuer if a shock is indicated. The typical AED weighs between 4 and 9 lb (1.8-4.0 kg) and costs between $3500 and $5000. Although AEDs vary in their mode of operation, the simpler models have 1-button technology and use voice prompts to instruct the rescuer what to do. Most AEDs store electrocardiographic and scene audio data for later review. The time required to train a rescuer to use an AED safely was originally specified as 4 hours. The actual time required to achieve proficiency may be as little as 1 hour.8

After several studies demonstrated that basic-level EMTs and first-responding firefighters could safely use an AED to treat ventricular fibrillation before paramedic arrival,9-11 the concept of "first-responder defibrillation" was promptly endorsed by the American Heart Association, the American College of Emergency Physicians, and the International Association of Fire Chiefs. In 1997, the American Heart Association recommended that every first responder be allowed and encouraged to use a defibrillator.12

Given the simplicity of AEDs, it was only a matter of time before the concept of "public access" defibrillation took hold. Within the past few years, AEDs have been placed in sports stadiums, hotels, shopping malls, businesses, casinos, airports, and airliners. In November 1998, the first person to be resuscitated by an AED on a domestic flight was reported.13 Recently, President Clinton announced that he has directed both the Department of Health and Human Services and the General Services Administration to create guidelines that will mandate placement of AEDs in all federal buildings as well as on all domestic and international flights.14

A search of the Lexis-Nexis database for a recent 6-month period (November 1, 1999–April 17, 2000) revealed 121 wire service and local news stories pertaining to AEDs. Not one was a negative report (Marlena Wald, MPH, MLS, written communication, April 18, 2000). The articles fell into several groupings: (1) "how an AED saved my life" human interest stories; (2) reports of rural areas that wanted placement of AEDs to deal with long ambulance response times; (3) stories about businesses purchasing AEDs to place in factories and other locations; (4) legislative pieces covering actions by state governments to release AED users from liability in the event that their "good samaritan" behavior is harmful; (5) fire and EMS agency lobbying to local governments for funds to purchase AEDs; and (6) the "even small children can operate an AED" story—a feature that has appeared in every region of the country, even though the report is based on a single, industry-sponsored study of 15 children and 22 adults.15

Lack of Evidence to Support "Universal Access" to AEDs

Despite widespread enthusiasm for this technology, there is little evidence to support the concept of "universal access" to AEDs. Researchers in Memphis, Tenn, conducted a large quasi-experimental trial of firefighter defibrillation. Forty engine companies of the Memphis Fire Department took part in the study; half were equipped with AEDs and half were instructed to perform CPR before paramedic arrival. Every 75 days, roles were reversed. Although first-responding engine companies arrived on the scene an average of 2.5 minutes prior to paramedic arrival, patients treated by an AED-equipped engine company were no more likely to survive to hospital discharge than those who received only firefighter CPR prior to paramedic arrival.16 A subsequent quasi-experimental trial conducted in Charlotte, NC, produced similar results.17

Other groups have reported more encouraging findings. Equipping police officers with AEDs in Rochester, Minn,18 and Pittsburgh, Pa,19 produced higher rates of survival of cardiac arrest compared with historical controls. Weaver and colleagues11 studied introduction of firefighter defibrillation in Seattle, Wash, and reported that the observed rate of survival for patients who had ventricular fibrillation or ventricular tachycardia was greater than expected based on historical data. It is important to note, however, that the overall rate of cardiac arrest survival in Seattle was not increased by widespread introduction of AEDs.

A meta-analysis of 7 studies of first-responder defibrillation detected an absolute risk reduction of 8.5% with the addition of manual or automated defibrillation.20 However, 4 of the studies were conducted in Seattle or in surrounding King County communities with unusually high rates of survival following out-of-hospital cardiac arrest.5,6,11,21 Other studies in cities with lower survival rates have failed to achieve similar results.16,17,22,23 Interestingly, the 3 trials that used contemporaneous control groups failed to demonstrate benefit.6,16,24 This suggests that at least some of the improvement noted in before-after trials of first-responder defibrillation was due to the Hawthorne effect.

The most recent study of the impact of introducing AEDs in the out-of-hospital setting was reported by Stiell and colleagues25 for the Ontario Prehospital Advanced Life Support (OPALS) study group. Nineteen urban and suburban communities in Ontario optimized their EMS systems by introducing firefighter defibrillation and other innovations to enable an AED-equipped team to reach the patient in 8 minutes or less in 90% of cardiac arrest cases. A total of 4690 patients with cardiac arrest were studied prior to system optimization; an additional 1641 patients were studied thereafter.

Following system optimization, the overall rate of survival to hospital discharge for all rhythms combined improved from 3.9% to 5.2% (P = .03). Surprisingly, this benefit was largely restricted to the subset of patients whose initial rhythm was pulseless electrical activity, a condition that is unresponsive to early defibrillation. The survival rate for patients found in the states of witnessed ventricular fibrillation or ventricular tachycardia (the rhythms most likely to benefit from early defibrillation) was not significantly improved (10% preintervention vs 11.9% postintervention; P = .17).

Based on the available data, it makes sense to provide AEDs to EMS services that currently lack the capacity to provide out-of-hospital defibrillation, but only if their ambulances can reach people with cardiac arrest within 8 minutes or less. Rural EMS services that have long response times are not likely to realize much benefit from carrying AEDs.24,26 It also makes sense to strategically place AEDs on city fire trucks that consistently reach people with cardiac arrest minutes before the nearest paramedic-staffed ambulance.6,10,16,21 It does not make sense to promote public access defibrillation when there is insufficient evidence that this strategy results in higher community rates of cardiac arrest survival.27 With rare exceptions, it is not even clear where public access AEDs should be placed because few locations are settings for more than 1 cardiac arrest per year.28

If applied in time, AEDs are clearly efficacious for terminating ventricular fibrillation or ventricular tachycardia. This does not, however, justify massive deployment of AEDs before it is shown to be a cost-effective strategy for reducing mortality from out-of-hospital cardiac arrest. The National Institutes of Health, the Agency for Healthcare Research and Quality, and other federal research agencies should actively solicit well-designed clinical trials to answer this question. Hopefully, some answers will be provided by the Public Access Defibrillation (PAD) Community Trial, which is funded by the National Heart, Lung, and Blood Institute and is currently under way in 25 locations across the United States.29 Until these data are available, a moratorium should be placed on establishing new public access defibrillation programs.

Some may argue that this level of caution is misplaced. After all, isn't doing something (purchasing large numbers of AEDs for distribution in a community) better than doing nothing, particularly for a condition as devastating as sudden cardiac arrest? In reality, the choice is more often between doing something and doing something else. The prices typically quoted for purchasing AEDs rarely include the hidden costs of initial courses and refresher training, electrodes, extra batteries, maintenance, and eventual replacement of obsolete units. Money spent on AEDs cannot be spent on other aspects of EMS operations, such as EMT training, airway adjuncts, supplies, medications, and vehicle maintenance. Little is gained (and much may be lost) if public access defibrillation programs siphon vital resources from aspects of EMS that benefit a wider range of patients.

No Justification for Over-the-Counter AEDs

Since public access defibrillation is already sweeping the country, can over-the-counter sales of AEDs be far behind? If wealthy people can afford to buy an AED for themselves, shouldn't they be allowed to? Hopefully, the answer to both questions is no. The fact that some people can afford to purchase a medical device that delivers a powerful shock does not absolve physicians of the responsibility to control access to it. Numerous medical tests, drugs, and devices pose a minimal risk of harm to patients, but nonetheless are subject to the professional discretion of a physician. A misused digital camera cannot cause harm; a misused AED certainly can.

More important, there is no evidence that placing an AED in the home of a patient with heart disease is better than teaching family members to immediately call 911 and begin CPR. As yet, the only published controlled clinical trial of AEDs placed in homes of cardiac patients produced negative, or, at best, inconclusive results.30 While there is little doubt that authorizing over-the-counter sales of AEDs would produce a financial windfall for the manufacturers of these devices, it is less clear that the public would benefit. Patients who face a particularly high risk of sudden cardiac death, such as those with refractory cardiac dysrhythmias and survivors of prior cardiac arrest, should be considered for an implantable defibrillator rather than an AED.

Clever salesmen might argue that purchasing an AED is equivalent to purchasing a home fire extinguisher—"It is better to have one and not need it than to need one and not have it." It is worth noting, however, that an AED is much more costly and requires more frequent checks and maintenance than a home fire extinguisher. The Consumer Product Safety Commission estimates that the smoke detectors in about 16 million homes do not work, mostly because the device's battery is dead or missing.31 What is the "shelf life" of an unused AED?

Early defibrillation is but one link in the chain of survival. Successful resuscitation after cardiac arrest depends on prompt recognition of the event, early initiation of CPR, rapid defibrillation, and early access to advanced cardiac life support. Encouraging over-the-counter sale of AEDs will do nothing to ensure prompt recognition of a cardiac arrest and may even delay notification of EMS and initiation of CPR if family members fumble with the device instead of calling 911.


The best way to reduce deaths from heart disease is through prevention. Given the choice of spending between $250 and $1500 to purchase an AED for the home or spending a comparable amount of money on a bicycle, a smoking cessation program, a health club membership, or treatment of hypertension, most people would be better served by choosing one of the latter options for themselves or a loved one.

Nonetheless, the power of marketing and the allure of medical technology should not be underestimated. Given the opportunity, it is a safe bet that many consumers will opt for an AED. After all, some might reason "Why should I exercise, stop smoking, and monitor my blood pressure when I can simply park an AED under my couch?"

Automated external defibrillators are wonderful devices, but they are not a panacea for cardiac arrest. Massive dissemination of this technology, including "over-the-counter" sale of AEDs, should not be considered unless this approach can be justified on the basis of sound scientific evidence. Otherwise, we may end up doing more harm than good.

Beck CS, Pritchard WH, Feil HS. Ventricular fibrillation of long duration abolished by electric shock.  JAMA.1947;135:985-986.Google Scholar
 Essentials of ACLS. In: Cummins RO, ed. Advanced Cardiac Life Support . Dallas, Tex: American Heart Association; 1997:1-7.
Liberthson RR, Nagel EL, Hirschman JC, Nussenfeld SR. Prehospital ventricular defibrillation: prognosis and follow-up course.  N Engl J Med.1974;291:317-321.Google Scholar
Cobb LA, Baum RS, Alvarez III H, Schaffer WA. Resuscitation from out-of-hospital ventricular fibrillation: four years follow up.  Circulation.1975;52:III223-III235.Google Scholar
Eisenberg MS, Copass MK, Hallstrom AP.  et al.  Treatment of out-of-hospital cardiac arrests with rapid defibrillation by emergency medical technicians.  N Engl J Med.1980;302:1379-1383.Google Scholar
Eisenberg MS, Hallstrom AP, Copass MK, Bergner L, Short F, Pierce J. Treatment of ventricular fibrillation: emergency medical technician defibrillation and paramedic services.  JAMA.1984;251:1723-1726.Google Scholar
Ornato JP, McNeill SE, Craren EJ, Nelson NM. Limitation on effectiveness of rapid defibrillation by emergency medical technicians in a rural setting.  Ann Emerg Med.1984;13:1096-1099.Google Scholar
Weisfeldt ML, Kerber RE, McGoldrick RP.  et al. for the Automatic External Defibrillation Task Force.  American Heart Association report on the Public Access Defibrillation Conference, December 8-10, 1994.  Circulation.1995;92:2740-2747.Google Scholar
Cummins RO, Eisenberg M, Bergner L, Murray JA. Sensitivity, accuracy, and safety of an automatic external defibrillator.  Lancet.1984;2:318-320.Google Scholar
Weaver WD, Copass MK, Hill DL, Fahrenbruch C, Hallstrom AP, Cobb LA. Cardiac arrest treated with a new automatic external defibrillator by out-of-hospital first responders.  Am J Cardiol.1986;57:1017-1021.Google Scholar
Weaver WD, Hill D, Fahrenbruch CE.  et al.  Use of the automatic external defibrillator in the management of out-of-hospital cardiac arrest.  N Engl J Med.1988;319:661-666.Google Scholar
Kloeck W, Cummins RO, Chamberlain D.  et al.  Early defibrillation: an advisory statement from the Advanced Life Support Working Group of the International Liaison Committee on Resuscitation.  Circulation.1997;95:2183-2184.Google Scholar
Associated Press.  In first, defibrillator revives man on US airliner.  New York Times.November 26, 1998:A22.Google Scholar
American College of Emergency Physicians.  AEDs now mandatory in airplanes, federal buildings.  EM Today.June 14, 2000:1-2.Google Scholar
Gundry JW, Comess KA, DeRook FA, Jorgenson D, Bardy GH. Comparison of naïve sixth-grade children with trained professionals in the use of an automated external defibrillator.  Circulation.1999;100:1703-1707.Google Scholar
Kellermann AL, Hackman BB, Somes G, Kreth TK, Nail L, Dobyns P. Impact of first-responder defibrillation in an urban emergency medical services system.  JAMA.1993;270:1708-1713.Google Scholar
Sweeney TA, Runge JW, Gibbs MA.  et al.  EMT defibrillation does not increase survival from sudden cardiac death in a two-tiered urban-suburban EMS system.  Ann Emerg Med.1998;31:234-240.Google Scholar
White RD, Asplin BR, Bugliosi TF, Hankins DG. High discharge survival rate after out-of-hospital ventricular fibrillation with rapid defibrillation by police and paramedics.  Ann Emerg Med.1996;28:480-485.Google Scholar
Mosesso Jr VN, Davis EA, Auble TE, Paris PM, Yealy DM. Use of automated external defibrillators by police officers for treatment of out-of-hospital cardiac arrest.  Ann Emerg Med.1998;32:200-207.Google Scholar
Auble TE, Menegazzi JJ, Paris PM. Effect of out-of-hospital defibrillation by basic life support providers on cardiac arrest mortality: a meta-analysis.  Ann Emerg Med.1995;25:642-648.Google Scholar
Weaver WD, Copass MK, Bufi D, Ray R, Hallstrom AP, Cobb LA. Improved neurologic recovery and survival after early defibrillation.  Circulation.1984;69:943-948.Google Scholar
Lombardi G, Gallagher J, Gennis P. Outcome of out-of-hospital cardiac arrest in New York City: the Pre-Hospital Arrest Survival Evaluation (PHASE) Study.  JAMA.1994;271:678-683.Google Scholar
Lui JCZ. Evaluation of the use of automated external defibrillation in out-of-hospital cardiac arrest in Hong Kong.  Resuscitation.1999;41:113-119.Google Scholar
Vukov LF, White RD, Bachman JW, O'Brien PC. New perspectives on rural EMT defibrillation.  Ann Emerg Med.1988;17:318-321.Google Scholar
Stiell IG, Wells GA, Field BJ.  et al.  Improved out-of-hospital cardiac arrest survival through the inexpensive optimization of an existing defibrillation program: OPALS Study Phase II.  JAMA.1999;281:1175-1181.Google Scholar
Gentile D, Auerbach P, Gaffron J, Phillips Jr J. Prehospital defibrillation by emergency medical technicians: results of a pilot study in Tennessee.  J Tenn Med Assoc.1988;81:144-148.Google Scholar
Riegel B. Training nontraditional responders to use automated external defibrillators.  Am J Crit Care.1998;7:402-410.Google Scholar
Gratton M, Lindholm DJ, Campbell JP. Public-access defibrillation: where do we place the AEDs?  Prehosp Emerg Care.1999;3:303-305.Google Scholar
Hallstrom AP. Public Access Defibrillation (PAD) Community Trial [abstract]. In: CRISP (Computer Retrieval of Information on Scientific Projects) Current Award Information [online database]. Grant 1N01HC95177. Available at: http://crisp.cit.nih.gov/. Accessed July 19, 2000.
Eisenberg MS, Moore J, Cummins RO.  et al.  Use of the automated external defibrillator in homes of survivors of out of hospital ventricular fibrillation.  Am J Cardiol.1989;63:443-446.Google Scholar
 CPSC warns the smoke detectors in about 16 million homes do not work [press release]. Washington, DC: Office of Information and Public Affairs, US Consumer Product Safety Commission; October 21, 1997.