Workplace safety
How workers die on the job in the EU: struck, crushed, machines
27.1% of fatal workplace accidents in the EU in 2023 came from losing control of a machine, tool, or transport equipment. A further 21.4% were workers struck by a moving object, and 13.8% were trapped or crushed.
Those three contact modes are exactly the ways a machine, including an industrial robot, kills: it moves unexpectedly, it strikes, or it traps. The figures are measured, from Eurostat's breakdown of how fatal accidents actually happen.
This page traces the contact-mode figures to their source and connects them to why machine guarding and lockout matter.
Data covers EU fatal accidents at work by contact mode, 2023 (Eurostat). Published 2026-07-20. Last reviewed 2026-07-20. Last updated 2026-07-20. Edited by Mike Ramsey / Reliable Media.
The figures and where they come from
Each figure is rated for how safely you can cite it today. Ratings judge current usability, not whether a number was ever correct.
| Figure | What it is | Source | Citation Confidence | Notes |
|---|---|---|---|---|
| 27.1% of deaths | Loss of control of machine/equipment | [A] | High | The single largest contact mode for EU workplace deaths in 2023: a machine, tool, or transport equipment going out of control. |
| 21.4% of deaths | Struck by a moving object | [A] | High | The second-largest mode: a worker hit by something in motion, including a machine's own movement. |
| 13.8% of deaths | Trapped or crushed | [A] | High | Workers caught, trapped, or crushed, the classic industrial-robot fatality mode when someone is inside a moving cell. |
Why the numbers disagree
Safety debate tends to focus on rare, dramatic events, but the data shows the deaths cluster in mundane mechanical modes. Loss of control of a machine (27.1%), being struck by a moving object (21.4%), and being trapped or crushed (13.8%) together account for well over half of EU workplace fatalities. These are equipment-motion deaths, not exotic ones.
That pattern is why machine guarding, interlocks, and lockout exist. An industrial robot kills in exactly these modes: it moves when someone does not expect it, it strikes, or it traps a person inside its envelope. The Eurostat categories are general, but they map directly onto the robot-cell hazard.
The categories describe the contact, not the root cause. A worker crushed by a machine may have been failed by a missing guard, a defeated interlock, or a skipped lockout, and the contact-mode figure does not capture which. It tells you how people die, not why the safeguard was absent.
How to cite these figures
Use the contact-mode split to ground machine-safety points: 27.1% of EU workplace deaths are loss of control of equipment, 21.4% struck by a moving object, 13.8% trapped or crushed.
Connect the modes to guarding and lockout, which exist precisely to break these contact paths, rather than treating machine deaths as freak events.
Keep contact mode separate from root cause. The figure says how a worker died, not whether a guard was missing or a lockout was skipped.
Where people go wrong
Treating machine fatalities as rare or exotic. The mundane mechanical modes account for most workplace deaths.
Reading a contact mode as a cause. Trapped or crushed describes the contact, not the missing safeguard behind it.
Assuming these EU categories are robot-specific. They cover all equipment; robots are one case that fits the same modes.
How we checked
The figures come from Eurostat's Statistics Explained analysis of the causes and circumstances of fatal accidents at work. We retrieved the article and confirmed the 27.1%, 21.4%, and 13.8% contact-mode shares in its text.
We connect the general categories to the robot-cell hazard because they map onto it directly, while being explicit that the categories cover all equipment, not robots alone.
We distinguish contact mode from root cause throughout, since the data records how a worker was killed, not whether a guard, interlock, or lockout failed.
Full source list
Primary sources, with live links. Every figure above traces to one of these.
- [A]Eurostat2025
Eurostat, "Accidents at work - statistics on causes and circumstances" (Statistics Explained)
https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Accidents_at_work_-_statistics_on_causes_and_circumstances
Common questions
- How do most workers die on the job in the EU?
- In mechanical contact modes. In 2023, 27.1% of EU workplace deaths came from loss of control of a machine or equipment, 21.4% from being struck by a moving object, and 13.8% from being trapped or crushed.
- How does this relate to robots?
- An industrial robot kills in exactly these modes: unexpected motion, striking, or trapping a person in its envelope. The Eurostat categories are general but map directly onto the robot-cell hazard.
- Does the data say why the safeguards failed?
- No. Contact mode records how a worker died, not whether a guard was missing, an interlock defeated, or a lockout skipped. Root cause is a separate question.
- Are these figures only about machines?
- No. They cover all fatal workplace accidents in the EU. Machine and equipment modes are the largest share, and robots are one case that fits them.
More data, traced to source
- Workplace deaths in the EU: 3,298 in 2023, and manufacturing's share
The EU recorded 3,298 workplace deaths in 2023, a rate of 1.63 per 100,000 workers. Manufacturing accounted for 443 of them and 522,000 non-fatal injuries. The measured safety record, from Eurostat.
- What robot fatality case files actually record
The individual robot-fatality investigations that do exist point to one failure again and again: the robot was live when it should have been locked out. What the OSHA and NIOSH-family case files record.
- Robot workplace deaths: what the public record does and doesn't show
The figure of about 41 US robot-related deaths is quoted widely, but it cannot be verified through any free primary source, and the same agency has published two different totals. This page traces what is and is not documentable about robot workplace fatalities.
- Why robot-death counts are overstated: the keyword trap
Counts of robot fatalities built by searching incident data for the word 'robot' overstate the toll. The database returns deaths that happened near a robot but were not caused by one, including a worker who died of natural causes in a robotic weld area.