Injury and illness data: where the numbers come from
Official injury and illness statistics come from different sources that answer different questions. You will read them, calculate a rate and say what the numbers leave out before you use them in a report.
- Optional
- Beginner
- about 5 hours
Written by the GreenHired editorial teamSources last checked 20 November 2026How we research roadmaps
Before you quote any figure, ask three questions: who counted, what counted as a case, and per what? The answers decide whether two numbers can sit side by side. Ask them in that order and write the answers beside the figure, because a manager who challenges your number will ask the same three things.
Which source answers which question
Each source answers one question well and others badly:
- How many people died at work in the US? The BLS Census of Fatal Occupational Injuries. Deaths are the hardest cases to leave out.
- How many non-fatal cases did employers record? BLS SOII in the US. In Great Britain, RIDDOR reports, which leave out some injuries such as most road traffic accidents.
- What do workers say about their own ill health? The Labour Force Survey, where people report it themselves.
- How do firms manage risk, as opposed to how many were hurt? EU-OSHA’s ESENER survey.
- How does my site compare with its industry? BLS industry rates and its incidence rate calculator.
Your own rates (TRIR, LTIFR) and their targets are built in safety performance metrics, and spreadsheets and dashboards turn the figures into charts a manager will read.
Site A against site B
Site A has 50 full-time workers and 3 recorded cases in a year, and site B has 500 full-time workers and 10. Both sites are invented, so the numbers are not real data.
Site B has more cases. Per 100 workers: A is 3 / 50 x 100 = 6.0, and B is 10 / 500 x 100 = 2.0. The smaller site has the higher rate. BLS builds its rates from hours worked, not headcount, so this shortcut only holds when everyone works full time all year.
Is B really three times safer?
Ask what counted as a case. If B pays a bonus for a low injury count, part of that 2.0 may be fewer reports, not fewer injuries. Put B’s near-miss reports, which come from incident reporting, and any serious cases next to the rate before you draw a conclusion.
What you will learn
- Where Great Britain, the US and the EU publish injury, illness and fatality data.
- How a rate differs from a count and why employers compare rates.
- What each data source includes and what it misses.
- Why non-fatal injuries are under-reported and how that distorts comparisons.
- How to use data in a short business case without overclaiming.
The parts of this topic
HSE statistics (Great Britain)about 1 hour
HSE combines employer reports under RIDDOR with the Labour Force Survey, so its figures come from two very different kinds of source.
HSE's data sources page explains that RIDDOR places a legal duty on employers to report certain incidents, but that some injuries, such as most road traffic accidents, are not reportable and so do not appear. The Labour Force Survey is a household survey in which people report their own work-related ill health and injuries. Reporting rules also change: in April 2012 the RIDDOR threshold moved from over-3-days to over-7-days off work. Note any such change before you compare a count from before 2012 with one after.
- HSE statistics: data sources HSE (UK) · Guide
- RIDDOR: reporting of injuries, diseases and dangerous occurrences HSE (UK) · Guide Landing page; the reportable-incidents pages hold the detail.
BLS SOII and CFOI (United States)about 1 hour
The US publishes two annual programmes: the Survey of Occupational Injuries and Illnesses for non-fatal cases and the Census of Fatal Occupational Injuries for deaths.
As read on 20 November 2026, the BLS IIF page reported 5,070 fatal work injuries for 2024 (CFOI) and about 2.4 million non-fatal cases for 2025 (SOII), with a private-industry incidence rate of 2.3 cases per 100 full-time equivalent workers. Check the page for the figures current when you read it. BLS also says it keeps respondents' data confidential and does not share any individual employer's rates, but it publishes industry rates you can compare against. Match the industry when you compare a company with a state or national figure.
- IIF questions and answers US Bureau of Labor Statistics · Guide
EU and global viewsabout 1 hour
Eurostat counts accidents at work across the EU, EU-OSHA surveys how firms manage risk, and WHO describes occupational health worldwide.
Eurostat's accidents at work statistics define a non-fatal accident as one causing at least four full calendar days of absence, and count fatal accidents that lead to death within one year. EU-OSHA's ESENER survey asks businesses and organisations across Europe how they manage safety and health risks in practice, which is a different question from how many people were hurt. WHO frames occupational health as protecting workers' health and working capacity. Do not treat the three as one dataset: each answers a different question about a different population.
- Occupational health World Health Organization · Guide
Incidence ratesabout 1 hour
A rate relates cases to hours worked or workers employed, so a small company and a large one can be compared fairly.
A count of injuries grows with the size of the workforce, so rates are used for comparison: ten recordable cases mean something different in a 50-person and a 5,000-person site. BLS publishes incidence rates per 100 full-time equivalent workers, and its online calculator asks for the hours worked at your establishment and your case counts, then compares you with national or industry figures. Rates used in your own reporting, such as TRIR or LTIFR, are covered later in the roadmap. Rates built on different definitions of a case cannot be compared.
- Injury and illness incidence rate calculator US Bureau of Labor Statistics · Tool
Under-reporting and its effectsabout 1 hour
Non-fatal injuries are reported unevenly, so a low number can mean good safety or weak reporting. Deaths are much harder to hide.
Eurostat notes that low non-fatal rates in some countries may reflect under-reporting through weak reporting systems or little financial incentive, and that insurance-based systems tend to show higher rates. Fatal accidents are harder to leave unreported, so they compare better. The same logic applies inside a company: a site with no reports may be the least safe, not the safest, and rewarding low numbers pushes reporting underground. Example: when a site starts a near-miss campaign, reported events can rise first.
- Accidents at work statistics Eurostat · Dataset
Using data in a business caseabout 1 hour
Managers act on cost, risk and comparison. Use official numbers, say what they cover and show the source so the case survives questions.
HSE's guidance for directors says good health and safety brings lower costs and risks, including fewer accidents, lower absence and turnover and less threat of legal action. A short case might compare your site's rate with the sector rate, show two incident costs you can evidence and propose one action with a price. A national total does not describe your site, so keep claims narrow: name the HSE table and year you used, not 'workplaces are dangerous'.
- Leading health and safety at work (INDG417) HSE (UK) · Guide Benefits, board duties and a checklist for leaders.
Practice task
Download one official dataset for your region (HSE, BLS or Eurostat). Write a one-page briefing with three findings about your sector, one sentence on what the source cannot tell you, and the exact table and date you used so someone else can check it.
Skills you build
Reading official statisticsIncidence ratesData limitsEvidence for business cases
Resources for this topic
Every source here is free. Each link opens the provider's own site in a new tab.
- Health and safety statistics (Great Britain) HSE (UK) · Dataset Open Government Licence; key figures, industries, costs and data tables.
- Injuries, Illnesses, and Fatalities (IIF) US Bureau of Labor Statistics · Dataset SOII and CFOI data, tables and tools.
- Facts and figures on occupational safety and health EU-OSHA · Dataset Includes the ESENER enterprise survey.
Live Environmental, health and safety (EHS) officer jobs
Roles whose title matches this career path.
Spot a problem with this topic?
Tell us what is wrong, missing or unclear here. Your note is attached to this topic.
Notes are sent from this page with JavaScript on. Without it, write to us on the contact page.