Elevator Accident Statistics: What the Government Data Shows

If You Were Hurt: What the Government Data on Elevator Accidents Shows

The most recent national figure comes from the CPSC's NEISS Data Highlights - Calendar Year 2024. The report's table measures estimated product-related injuries "in the United States and Territories" that "Were Treated in Hospital Emergency Departments." The row labeled "Elevators, Other Lifts" shows these figures for 2024:

• The estimated number of injuries is 31,938. The estimated rate is 9.5 per 100,000 population.
• The report's coefficient of variation, a measure of sampling variability, is .18.
• The actual number of injury cases collected from the sampled hospitals is 822.
• By age, the estimates are 1,259 for ages 0–4, 683 for ages 5–14, 1,406 for ages 15–24, 11,981 for ages 25–64 and 16,557 for age 65 and older. The rate for the 65-and-older group is 27.1 per 100,000.
• By sex, the estimates are 12,980 male and 18,930 female.
• By disposition, 26,144 were in the column the report labels "Treated & Rel." and 5,794 were in the column it labels "Hosp. & DOA."

The report attaches its own caveats. Because NEISS is a probability sample, each case "has a statistical weight," and the estimates "are derived by summing the statistical weights." A person's injury "may be counted in three product groups," because each injury can be reported under up to three products.

The second figure is older and narrower. In a June 24, 2021 news release, the CPSC stated that "residential elevators were linked to 4,600 injuries and 22 deaths from 1981 through 2019." It is a historical total for home elevators over 39 years, not a yearly count. People searching for elevator deaths per year will not find an annual death total for all elevators in either document.

People searching for escalator accident statistics should note that the 2024 report's category reads "Elevators, Other Lifts." The report itself defines what falls under that label. Neither figure is a finding about any company, building, product or operator.

For someone who was hurt, what is at stake is mostly clarity: these figures show that elevator injuries are tracked nationally, but they say nothing about whether your own injury was preventable. A person in this position may weigh which figure, if any, matches their setting, a home elevator or a commercial building, and may keep these national numbers separate from the records of their own incident.

How National Figures Relate to Your Own Accident Records

This page shows a person hurt in an elevator or escalator accident, or a family after a death, what the federal government's own published figures say, in the words those reports use. A figure drawn from a primary source can be checked against the report itself, unlike a number repeated across websites.

People who come to this page often already hold records of their own: an incident report the building made, photos of the elevator car or the landing, emergency room and follow-up medical records, and the names of people who saw what happened. National statistics sit beside those records. They do not replace them. A figure about 31,938 estimated emergency department visits says nothing about the elevator in your building, its maintenance contractor (the company hired to service and repair it) or the moment the doors closed.

The data cannot say what caused an accident, who is responsible, whether a claim exists, which deadline applies or what any case may involve. Those questions turn on the facts of one incident and on the law of the state where it happened. That is where the stake lies for a reader: the answers come from your own records, not from national totals. A person may decide to gather those records in one place, and a conversation with an attorney can help sort out what they show.

Where These Figures Come From and Why the Source Matters to You

Both figures on this page come from the CPSC, the federal agency that describes itself as "charged with protecting the public from unreasonable risk of injury associated with the use of thousands of types of consumer products." For a reader, that means these numbers describe consumers using products, not workers on the job.

The 2024 figure comes from the National Electronic Injury Surveillance System, known as NEISS. The CPSC describes NEISS as a system that "collects current injury data associated with consumer products from U.S. hospital emergency departments across the country." It consists of "a national probability sample of hospitals of differing sizes and locations." The 2024 estimates "have been produced from a sample of 100 hospitals, of which 77 reported in 2024, including children's hospitals." Rates use the Census Bureau's July 1, 2024 U.S. resident population estimates. The source is the NEISS 2024 Data Highlights report, page 5.

The 1981–2019 figure comes from the CPSC's news release on residential elevators, issued June 24, 2021 as Release Number 21-153. The release focuses on "a deadly gap that may exist between the elevator door and the exterior (i.e., hoistway) door inside the home." The hoistway is the shaft the elevator car travels in. The release states that "Children, some as young as two, up to age 16, have been crushed to death in this gap." For a family whose child was hurt in a home elevator, this is the federal source closest to that situation. The release does not describe the method behind its multi-decade total in the passages quoted above, so this page reports the figure only as stated.

If you were hurt at work in an elevator, a different set of agencies is relevant. The Occupational Safety and Health Administration (OSHA) and the Bureau of Labor Statistics (BLS) deal with workplace safety and work-related injuries and deaths, and the National Institute for Occupational Safety and Health (CDC/NIOSH) conducts occupational health research. This resource does not draw figures from those agencies, and it does not state any.

Each source counts a different group in a different setting: NEISS covers people of all ages treated in emergency departments in one product category, the 2021 release covers residential elevators only, and workplace data covers workers. A family looking at a figure about an office tower, a mall escalator or a home elevator may ask which of these sources the figure actually came from before relying on it.

Interpretation

How to Read an Injury Estimate: What the Numbers Mean

An estimated injury figure is a national projection built from weighted hospital cases, as described above, not a list of every incident. The 31,938 figure is an estimate of emergency-department-treated injuries in the "Elevators, Other Lifts" category for 2024. It does not count every injury or cover every setting.

Two other numbers in the same row show how firm that estimate is. The raw count of 822 is "The actual number of injury cases collected from the hospitals sampled." The report warns that because cases carry different weights, "these 'raw' numbers should not be used for comparative purposes." The coefficient of variation of .18 is "the standard error of the estimate divided by the estimate," which the report calls "a measure of sampling variability (errors that occur by chance because observations are made only on a population sample)." Many rows on the same page show a CV of .08 to .12. A higher CV means more sampling uncertainty. For a reader, this means the elevator figure is a rough national picture, and any precise claim built on it deserves a second look.

The 2021 figure's 4,600 injuries and 22 deaths are spread across 1981 through 2019 and limited to residential elevators. Dividing that total by the number of years would produce a number the CPSC never published, so this page does not do it.

The phrase "linked to" describes an association recorded in the agency's data. It is not a determination of what caused any particular injury or of anyone's fault.

Some publications report elevators and escalators together, and some separately. Escalator figures from one source and elevator figures from another may rest on categories defined differently.

Questions Worth Asking Before Drawing Conclusions from Elevator Injury Data

When a figure comes up, whether in an article, a conversation or a letter from an insurer, a few questions help place it:

1. What population does the figure cover? It may count consumers, workers or all persons, and residential, commercial or all elevators.
2. What time period does it cover? It may be a single year, a multi-year total or a multi-decade total such as 1981 through 2019.
3. What data source produced it? It may be a NEISS hospital-based estimate, workplace fatality data from OSHA or BLS, or another source.
4. What is its unit? It may be estimated injuries treated in emergency departments, reported deaths or something else.
5. Does it include escalators, moving walkways or other lifts, or only elevators?

These questions do not change what happened to you. They do help show whether a number someone cites actually fits the situation it is being used to describe.

Limitations

Why the Statistics Do Not Tell the Whole Story of Any One Incident

The NEISS estimate covers injuries treated in hospital emergency departments. An injury treated only at an urgent care clinic or a doctor's office falls outside it. So does an injury that was never treated, or one coded to a different product category. If you were hurt and went to your own doctor days later, your injury may not appear in this data at all. It is still real, and it is still recorded in your own medical records.

A statistic also cannot establish cause. In general terms, many kinds of failures can lead to elevator accidents. Entrapment means being stuck inside a stopped car. Misleveling means the car stops above or below the floor, which can cause a trip or fall. Door failures and sudden drops are other examples. What caused any one incident depends on its facts: the equipment, its condition and what happened that day.

Injury statistics are also separate from inspection and maintenance records. A figure about how often injuries occur says nothing about whether a particular elevator was inspected, serviced or in compliance with the codes that apply to it. An inspection or maintenance record is not an allegation or a finding of negligence.

The law also varies. Limitation periods, which are the time limits for filing a claim, differ from state to state, as do fault rules and damages rules. A national statistic does not determine which state's law applies to anyone's claim. One state's approach is set out on the site's California elevator accident law page. California's rules are California's alone.

This resource never determines fault, liability, a deadline or what a case is worth.

What Records Often Matter When Someone Is Hurt in an Elevator

For a person who was hurt, the record that usually counts most is the record of their own incident. The records that usually matter include:

• An incident report made with the building owner or operator
• The elevator's inspection certificate and inspection history
• Maintenance and service logs kept by the owner or its maintenance contractor
• Surveillance or security camera footage
• Medical records documenting the injury and treatment
• The names of witnesses

Inspection requirements and recordkeeping rules vary by state and by locality. No single national rule governs every elevator. An inspection record shows whether equipment was examined, when and by whom. A finding of negligence is a legal conclusion about whether someone failed to use reasonable care. The two are separate. Premises liability, the law about a property owner's duties to people on the property, and product liability, the law about defective products, each ask their own questions, and whether they are met depends on the facts.

Pages on specific companies, such as Otis Elevator Company and Schindler elevator injury claims, describe what official records state about them and nothing more. A person sorting through an incident report and medical records may want to consider talking through those records with our attorney, including what the official record of the incident shows and how it fits with the rest.