Reference: JAMA Intern Med. 2026 Jan 1;186(1):37-43
Practice Point: Until someone makes a less expensive automated external defibrillator (AED), the improved survival associated with in-home AEDs is not cost-effective enough to recommend widespread implementation.
EBM Pearl: A difference-in-difference study design strengthens causal relationships in observational data by using an internal negative control to factor out any shared confounding.
Out-of-hospital cardiac arrest remains one of medicine’s most challenging emergencies. It affects more than 300,000 people annually in the United States alone, with fewer than 10% of patients surviving to hospital discharge. Compared with events in public settings, home arrests are less likely to be witnessed and less likely to involve a shockable rhythm. At the same time, patients are less likely to receive bystander cardiopulmonary resuscitation (CPR) and more likely to experience longer emergency medical service (EMS) response times. These factors make private homes one of the worst locations for cardiac arrest.
Publicly placed AEDs are among the rare interventions in resuscitation medicine that appear to deliver on their promise. They may improve survival in patients with shockable rhythms and are cost-effective when deployed in accessible locations, such as the repurposed red phone booths now housing AEDs throughout the United Kingdom. Whether that success translates to private homes has been far less clear.
Using the Cardiac Arrest Registry to Enhance Survival, investigators analyzed more than 582,000 nontraumatic cardiac arrests occurring in private homes between 2017 and 2024. Only 16% of patients presented with an initial shockable rhythm, and AEDs were applied by bystanders in just 0.1% of cases. As randomized trials at this scale are not feasible, the authors used a difference-in-difference design that leverages one truth: AEDs cannot improve outcomes in cardiac arrests with nonshockable rhythms.
A difference-in-difference analysis asks whether outcomes improve more than expected in a group that might benefit from an intervention compared with a group that should not. Patients with shockable rhythms serve as the exposed group, while those with nonshockable rhythms act as a negative control. Changes seen in both groups reflect background trends in resuscitation care, whereas any additional improvement confined to shockable rhythms is more likely attributable to AED use rather than confounding. This approach strengthens causal inference in observational data by comparing changes over time and accounts for baseline differences and confounding variables.
Among patients with shockable rhythms, AED application was associated with improved survival to hospital discharge (risk ratio 1.26 [95% CI 1.01-1.579]) and higher rates of favorable neurologic outcomes (risk ratio 1.33 [95% CI 1.05-1.67]). The adjusted difference-in-difference analysis for shockable rhythms also reported a survival benefit (risk ratio 1.26 [95% CI 0.82-1.95], although the lack of statistical significance could be limited by low event rates. No benefit was observed in patients with nonshockable rhythms, which strengthens the argument that the observed survival advantage is causal rather than confounded. The magnitude of effect is consistent with prior studies of AED use in public settings.
However, effectiveness does not necessarily equal value. When pairing their clinical analysis with a cost-effectiveness model, the authors’ results show that placing a $1,620 AED in private homes costs approximately $4.5 million per quality-adjusted life-year gained. This figure is driven by very low event rates, given that the annual incidence of cardiac arrest in a private home is 0.05% and that most arrests are not shockable. According to the model, AEDs would need to cost roughly $65 over their usable lifespan to be cost-effective.
No study of this scale is bias-proof, and selection bias may be unavoidable in this case. Households with AEDs are likely wealthier, more health-literate, and more prepared to intervene effectively. The difference-in-difference approach helps mitigate this, but it cannot eliminate all bystander-level confounding. Misclassification of initial rhythm is another concern, particularly if shockable rhythms deteriorate before EMS arrival, potentially diluting observed effects.
AEDs undoubtedly save lives when shockable rhythms are present, even in private homes. But at current prices, routine home AED purchases are not cost-effective and should not be broadly recommended. The highest-value interventions remain early recognition, bystander CPR, rapid EMS response, and strategic AED placement where cardiac arrests actually happen. Evidence-based medicine doesn’t always give us comforting answers, but it reliably keeps us honest.
For more information, see the topic Adult Basic Life Support (BLS) in DynaMed.
DynaMed EBM Focus Editorial Team
This EBM Focus was written by Rich Lamkin, MPH, MPAS, PA-C, Clinical Writer at DynaMed. Edited by Katharine DeGeorge, MD, MS, Executive Editor at DynaMed and Associate Professor of Family Medicine at the University of Virginia; Alan Ehrlich, MD, FAAFP, Executive Editor at DynaMed and Associate Professor in Family Medicine at the University of Massachusetts Medical School; McKenzie Ferguson, PharmD, BCPS, Senior Clinical Writer at DynaMed; Claire Symanski, PhD, Medical Editor and Team Lead for ENT at DynaMed; Michael Butler, PhD, Medical Writer at DynaMed; Matthew Lavoie, BA, Senior Medical Copyeditor at DynaMed; Hannah Ekeh, MA, Senior Associate Editor II at DynaMed; and Jennifer Wallace, BA, Senior Associate Editor at DynaMed.