Industries / Government, education and community / Government / Emergency services / Ambulance and medical transport
Vehicles and equipment
Building the ambulance itself and the stretchers, chairs and devices crews use to lift, carry and treat patients on the way.
6
Milestones achieved, from 1908 to 2018.
2
To watch for: announced, not yet confirmed outside.
3
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1908
Motors replace the horse
New York City's first motor ambulances arrived in 1908. Steam, electric and gasoline vehicles competed for the city's business, and gasoline won because it proved the most reliable. The city retired its last horse-drawn ambulance in 1922.
Lessonthe city chose the engine that broke down least, and even then the change took 14 years.
Credit
New York City's hospital ambulance services
1974
One specification for America
On 2 January 1974 the US General Services Administration (GSA), which buys for federal agencies, issued KKK-A-1822, a purchasing specification for ambulances. Its 1980 revision set three body types: Type I, a truck cab with a separate box body; Type II, a van; and Type III, a van cab with a box body. It required a patient compartment at least 116 inches (2.95 m) long and 60 inches (1.52 m) high, and from 1981 a maker's certificate and label before an ambulance could carry the "Star of Life" symbol. Because it was the only document available, local, state and federal buyers using federal grants adopted it, and it became the de facto US standard.
Credit
GSA's Federal Supply Service; the National Highway Traffic Safety Administration (NHTSA), which approved its test parameters
2014
Machines press no better
A mechanical chest compression device pushes on the chest of a person whose heart has stopped, at a steady rate and depth, in place of a crew member doing cardiopulmonary resuscitation (CPR) by hand. In the LINC trial, six ambulance services in Sweden, Britain and the Netherlands treated 2,589 patients from 2008 to 2013 with either the LUCAS device or manual CPR. Four hours later 23.6% of the machine group and 23.7% of the manual group were alive. A 2015 pooled analysis of five trials of the LUCAS and AutoPulse devices found no survival advantage either.
Lessona machine that never tires did not save more lives than trained hands.
Credit
Sten Rubertsson and colleagues (LINC); Simon Gates, Gavin Perkins and colleagues (pooled analysis)
2017
Crash tests for the back
Cars must meet federal crash standards, but in the US the safety of an ambulance's patient compartment is left to each state. NIOSH, the federal worker safety institute, and the Department of Homeland Security crash-tested ambulances in front, side and rear impacts with more than 30 partners, among them ambulance builders and cot and seat suppliers, and SAE International published 10 test methods covering crew seats, cots and their floor mounts, equipment mounts, cabinets and the box body. NIOSH cites an NHTSA review of 45 crash investigations from 2001 to 2015 that found 84% of crew members in the patient compartment were not wearing a seat belt. The 2016 edition of NFPA 1917, the National Fire Protection Association's ambulance standard, and the Ground Vehicle Standard of CAAS, the ambulance accreditation commission, were set to require the seat, cot and equipment-mount tests, and a July 2017 change notice to the federal specification added four more of the methods.
Lessona crash test protects crews only where a state's rules or a buyer's order require it.
Credit
NIOSH (Jim Green and colleagues); the DHS Science and Technology Directorate; SAE International; ambulance, seat and cot makers
2017
Powered stretchers pay back
A powered stretcher raises and lowers the patient by motor, and a power-load system lifts the loaded stretcher into the ambulance. After Niagara Emergency Medical Service in Ontario brought in both, stretcher-related muscle and joint injuries fell from an average of 20.0 per 100 full-time staff a year, over four years, to 4.3 in the year after, a drop of 78%. At Hamilton Paramedic Service, the comparison service, the rate rose from 17.9 to 24.6. The authors estimated that lower compensation costs would repay the equipment within its 7-year service life. In Ohio, 297 grants from an insurance safety program paid $16.67 million (in 2018 dollars) for powered cots, loaders and stair chairs from 2005 to 2018; the services reported fewer injuries, and the heavier cot was the main complaint.
Credit
Niagara EMS; Daniel Armstrong, Steven Fischer and colleagues (Ontario study); M. A. Hayden and colleagues (Ohio review)
2018
Thirty-two kinds of ambulance
A September 2018 review for NHS Improvement, led by Lord Carter, found that England's ambulance trusts ran over 5,000 emergency vehicles, including 3,200 double-crewed ambulances in 32 different designs, with no standard list of what each carried. All ran on diesel, and the average ambulance travelled about 50,000 miles (80,000 km) a year. The review asked for one national specification and joint buying. By 2025 the East of England Ambulance Service described anti-idling technology on its new diesel ambulances as standard "as part of the national ambulance specification".
Lessoneach trust had been paying its own price for the same conversion work.
Credit
Lord Carter; NHS Improvement
Watch for
London's electric ambulances
London Ambulance Service says it now runs 41 fully electric ambulances, the UK's largest such fleet, after adding 37 in summer 2026, and that they cover a full 12-hour shift on one charge and spend 75% less time off the road for repairs. Its first electric ambulance began answering 999 calls on 31 December 2023 as one of 12 that NHS England's Net Zero Travel and Transport team planned to pilot with five trusts to collect "real-life" data.
Achieved whenNHS England or an outside evaluator publishes the pilot's measured range, energy use, winter performance and time off the road against diesel ambulances.
Claimed by
London Ambulance Service, July 2026
A year on battery power
Hato Hone St John, which runs emergency ambulances for 90% of New Zealand's people by its own count, reports that its electric Ford E-Transit ambulance in Hamilton attended 521 incidents and travelled 13,671 km in its first year. It had to be swapped for a diesel ambulance twice, both times for process reasons, and used about 0.33 kWh per km by the report's own table. The service says it can complete a 12-hour city shift, and has moved the vehicle to Lyttelton to test colder, hillier ground.
Achieved whenan outside party publishes the results, including the colder second phase, or the service reports the same figures across several electric ambulances.
Claimed by
Hato Hone St John, March 2026
Open
01
State ambulance rules, mapped
A public table, for every US state, the District of Columbia and the five inhabited territories, of which ambulance design standard its rules require: the federal KKK-A-1822F, NFPA 1917 or its 2024 successor NFPA 1900, the CAAS Ground Vehicle Standard, or none, and which of the SAE crash test methods each rule brings in. Each cell links to the rule text and its effective date. Rerun it every year.
The briefCredit · unclaimed02
Range on a cold shift
A repeatable test of a converted electric ambulance driving a fixed 12-hour duty cycle on a closed track or private road, with a weighted adult manikin on the cot, a full equipment load and the patient compartment held at 21 °C while parked at a mock hospital. Run it at three outside temperatures, below −5 °C, near 5 °C and above 20 °C, and publish energy per km, power drawn while parked and the charge left at the end of each shift. The vehicle is a demonstrator or a retired unit, never one in service.
The briefCredit · unclaimed03
Reach from the seat, measured
A published audit of current production ambulances against the reach rule in ambulance ergonomic standards: from a belted seat, a small attendant and a large one must reach a belted patient from the crown of the head to the kneecap, and reach the main controls and the sharps bin. Score parked demonstrator vehicles at dealers and trade shows with volunteers near the 5th percentile female and 95th percentile male body sizes and a full-size manikin on the cot.
The briefCredit · unclaimed
Open 01 · the brief
State ambulance rules, mapped
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table, for every US state, the District of Columbia and the five inhabited territories, of which ambulance design standard its rules require: the federal KKK-A-1822F, NFPA 1917 or its 2024 successor NFPA 1900, the CAAS Ground Vehicle Standard, or none, and which of the SAE crash test methods each rule brings in. Each cell links to the rule text and its effective date. Rerun it every year.
Why it is still open
The latest counts found are a decade old. In January 2015 the head of the National Association of State EMS Officials (NASEMSO) said 42 states had rules on ambulance design and 30 relied on KKK-A-1822F, and a March 2015 status paper said no state had yet adopted another standard. Since 1 January 2024 NFPA 1917 has been folded into NFPA 1900, so state rules that cite 1917 may point at a replaced document. NASEMSO's state-by-state charts no longer load at their web addresses (checked November 2026), and a search that month found no current survey.
Built with what exists
state administrative codes and EMS rule pages; a browser agent such as Browser Use to collect rule text from state sites (see the Note "Deleting 80% to start over"); a general AI model to pull the cited standard and test methods into fixed fields, with two people checking each one; a public spreadsheet or CSV in a code repository.
First step
Read the ambulance vehicle rules of 10 states and record the standard and the SAE test methods each one cites.
Done looks like
all 56 jurisdictions covered, every cell linked to rule text with its date, two reviewers agreeing on at least 95% of cells in a 20% sample, and a second yearly update published.
Open 02 · the brief
Range on a cold shift
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A repeatable test of a converted electric ambulance driving a fixed 12-hour duty cycle on a closed track or private road, with a weighted adult manikin on the cot, a full equipment load and the patient compartment held at 21 °C while parked at a mock hospital. Run it at three outside temperatures, below −5 °C, near 5 °C and above 20 °C, and publish energy per km, power drawn while parked and the charge left at the end of each shift. The vehicle is a demonstrator or a retired unit, never one in service.
Why it is still open
The one published year of in-service data, from Hato Hone St John in Hamilton, New Zealand, comes from a mild city, and its colder second phase has not reported. NHS England's 12-vehicle pilot has not published the data it set out to collect. Planning models run on assumptions: ELASPY, an open-source simulator for electric ambulance fleets, uses 0.4 kWh per km driving and 5 kW while parked, values "provided by industry experts", and a 2026 analysis of US national ambulance records tested two hypothetical vehicles. A search in November 2026 found no cold-weather test of an electric ambulance.
Built with what exists
a demonstrator electric ambulance from a converter or dealer; an OBD or CAN bus data logger that reads the vehicle's own state of charge and power draw; calibrated thermometers inside and out; a weighted adult training manikin; the ELASPY simulator, to turn the measured figures into effects on response times for a region.
First step
Write the duty cycle from the times on task and trip distances in Maher and colleagues' 2026 analysis of US records, and ask two ambulance converters for a demonstrator.
Done looks like
the same duty cycle driven three times at each of the three temperatures, energy per km and parked power reported with their spread, all logs published, and ELASPY rerun for one region with the measured values in place of the assumed ones.
Open 03 · the brief
Reach from the seat, measured
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A published audit of current production ambulances against the reach rule in ambulance ergonomic standards: from a belted seat, a small attendant and a large one must reach a belted patient from the crown of the head to the kneecap, and reach the main controls and the sharps bin. Score parked demonstrator vehicles at dealers and trade shows with volunteers near the 5th percentile female and 95th percentile male body sizes and a full-size manikin on the cot.
Why it is still open
Canada's CSA D500:20 ergonomic standard for ambulances and a 2015 guidebook from the US Department of Homeland Security both set this rule. The guidebook says compartments are "typically not designed" for crews to work seated and belted, and NHTSA found 84% of crew members in the back unbelted in 45 crashes. Work on reach so far models designs before they are built, such as a 2013 NIST simulation of care tasks. A search in November 2026 found no published measurement of production ambulances against the rule.
Built with what exists
published body-size tables to recruit volunteers; a tape measure and a marked reach board; an adult training manikin; phone video with an open pose-tracking model such as MediaPipe to log each reach; the guidebook's own checklist.
First step
Score three demonstrator ambulances at one dealer with two volunteers and the guidebook's checklist.
Done looks like
20 current models from at least five builders, each scored on at least six reach points for both body sizes by two raters, with photos and a pass or fail table per point.