Drone-delivered AEDs.
A field guide.

Start with the response problem

An automated external defibrillator arriving by drone is a compelling idea. The useful question is whether a particular community can get an AED into a bystander’s hands sooner, reliably enough to justify the work and cost. I would evaluate the full response: dispatch, delivery, retrieval, use, and the ambulance that is still on its way.

This guide separates reported results from the questions I would ask when assessing a pilot. It is a starting point for commercial and operational planning, not a claim that any particular aircraft, location, or business has permission to fly.

Real programs, with local limits

Everdrone reports emergency medical operations in Sweden and France. In Florida, Manatee County describes an emergency drone program with ArcherFRS and Tampa General Hospital. Duke reports drone AED dispatch during real 911 calls in Forsyth County, North Carolina, as part of a study that also includes James City County, Virginia, supported by the American Heart Association.^[1][2][3]

These are useful examples to study. They do not establish that the same operating model, coverage, or response advantage transfers to another community. I would ask a prospective partner to distinguish what is operating now from what is proposed, and to show the conditions under which a delivery can actually happen.

Minutes gained are not the same as outcomes proved

In a Swedish series covering 2021–22, drones completed 58 AED deliveries. Among deliveries with comparable arrival-time data, the drone arrived before the ambulance in 67% of cases, with a median advantage of 3 minutes 14 seconds. Among confirmed arrests where the drone arrived first, an AED was attached in 6 of 18 cases; one patient survived to 30 days.^[4]

The contrast matters. In Aalborg, Denmark, the median time from activation to drone AED delivery was 4:47, compared with 3:25 for the ambulance. No delivered AED was attached. The studies support feasibility and show why location matters. They do not establish a general survival benefit from adding drones.^[4][5]

Flight permission is part of the product

FAA and PHMSA guidance issued in November 2025 explicitly includes lithium batteries inside AEDs among hazardous materials. Ordinary Part 107 operations prohibit carrying hazardous materials; that prohibition cannot be waived through the ordinary waiver process. A first-responder flight waiver should not be treated as permission to carry an AED. An operator may need a separate FAA exemption or another authorization route; those permissions must cover the specific operation and payload.^[6][7]

As of this guide, the proposed Part 108 framework for beyond-visual-line-of-sight operations remains under White House review. I would require the proposed operator and public-agency partner to establish the applicable flight and payload permissions before committing to a launch. A proposed rule is not an operating approval.^[8]

Name the payer and the continuing obligations

Hospital sponsorship is a demonstrated funding model: Tampa General Hospital sponsors the Manatee County program. Research support is another model, as the American Heart Association’s backing of the Duke study shows. Neither example is a promise that a new project will receive funding.^[2][3][9]

My planning brief would name the payer, the operator, the agency responsible for emergency response, and the owner of the equipment. It would cover recurring service, maintenance, training, reporting, and what happens if the sponsor leaves. A successful demonstration and a sustainable service are different decisions.

Choose the gap, then choose the equipment

I would start with local response records and the locations of accessible AEDs. An exurban fire district with persistent delays may deserve closer study than a place where ambulances consistently arrive quickly. That is a site-selection hypothesis to test, not a guarantee about rural or urban performance.

I would also compare a drone proposal with improving ground access: making existing AEDs easier to find, keeping them ready for use, and extending access beyond building hours. A pilot brief should explain what delivery adds to those options and how maintenance, training, and funding would work.

The decision I would put in front of a team is practical: where is the access gap, which approach addresses it, who will run it, and what evidence would justify continuing? The aircraft is one part of that answer. The service around it is the rest.