What the statistics tell us about sawmill risk

A sawmill puts heavy moving material, powerful cutting tools, long conveyor lines, fine combustible dust and constant noise under one roof. This first lesson looks at what open statistics say about accidents and, just as important, how far those numbers can be trusted and compared.

The EU picture (Eurostat, 2024)

Eurostat's ESAW (European Statistics on Accidents at Work) is the common reference for EU-wide comparisons. For 2024 it reports:

Fatal accidents were 45 (1.4%) more than in 2023, while non-fatal accidents were 43,742 (1.5%) fewer. Worked example: the 2023 fatal total was therefore 3,367 - 45 = 3,322.

Manufacturing, the sector that contains sawmilling, has the largest share of non-fatal accidents. Be careful, though: the Eurostat article does not break out wood or furniture manufacturing, so there is no EU wood-sector rate in the portal's source. Never quote one as if it existed.

What kinds of events dominate (EU, 2023, all sectors)

These are all-sector figures, not sawmill figures. The portal's reading (an interpretation, not a statistic) is that losing control of a tool, being struck by something moving and being trapped or crushed describe in generic terms what happens when a hand meets a blade, a log rolls or a person is drawn into a conveyor.

Country snapshots

• Germany (DGUV, 2025): 730,573 reportable work accidents and 328 fatal work accidents across the insured economy; the sawmill page of the DGUV wood and metal committee says that accidents at sawmill machinery often cause very severe injuries.
• Poland (GUS, 2024, preliminary): manufacturing had 7.56 injured per 1,000 workers, the third-highest sector rate. These data were preliminary, so treat them as provisional.
• Poland, wood processing 2010-2014 (reported from the labour inspectorate PIP via a newspaper): 508 injured, 51 killed and 257 seriously injured in wood-processing plants. This is old and second-hand; use it for pattern recognition, not as a current rate.
• United Kingdom (HSE; not EU): the sawmilling major injury rate is described as over two and a half times that of general manufacturing. The page is undated, so it is qualitative.

Why you cannot compare the numbers directly

Eurostat, DGUV, GUS, PIP and HSE use different definitions, years and denominators (per 100,000 employed, per 1,000 full-time equivalents, per 1,000 workers). A German figure of 16.68 per 1,000 full-time equivalents and a Polish 4.80 per 1,000 workers are not a league table. Read each number for its own country, year and definition.

The recurring pattern

Across countries the descriptive pattern is similar: severe cases cluster around machinery intervention (clearing, cleaning, adjusting), material handling and stacking, and energy control (machines that move when someone believes they are stopped). The Polish inspection campaign found typical management causes: economics placed ahead of safety, safety services that existed on paper, limited knowledge of hazards, weak supervision with no systematic checks of machine maintenance, and tolerance of hazards.

Checklist: using a statistic responsibly

1. Which country or region is it from?
2. Which year, and is it final or preliminary?
3. Which sector and which denominator?
4. Is it a primary source or a newspaper report of one?
5. Am I comparing it only with a number defined the same way?

Key points

• EU 2024: about 2.78 million non-fatal and 3,367 fatal accidents at work; manufacturing is 18.3% of non-fatal accidents.
• No EU wood-sector accident rate is available from the portal's source; do not invent one.
• Loss of control of machines, being struck and being trapped or crushed are major event types (EU, 2023, all sectors).
• National statistics use different definitions, years and denominators and must not be compared directly.
• Severe sawmill accidents cluster around machinery intervention, handling and stacking, and uncontrolled energy.