Published 8 October 2026. All figures are from the Health and Safety Executive. Fatality data is provisional for 2025/26 and will be finalised in July 2027. Free to cite with attribution — see how to cite this page.
Key findings
- 126 workers died in work-related accidents in Great Britain in 2025/26 — a rate of 0.37 per 100,000 workers, down from 128 deaths and 0.38 the year before. A further 104 members of the public were killed.
- Falls from a height caused 31 of those deaths — 24.6% of the total, and more than twice as many as the next single cause. Contact with moving machinery caused 10.
- 680,000 workers were injured in 2024/25 on HSE's survey estimate, but employers reported only 59,219 injuries under RIDDOR — about 8.7% of the total.
- The deadliest industry and the most injurious industry are not the same. Construction recorded the most deaths (25); accommodation and food services has the highest non-fatal injury rate (3,080 per 100,000 workers), and does not appear in the fatality table at all.
- Wales had the highest fatality rate of any nation or region at 0.62 per 100,000 workers — nearly five times London's 0.13. Scotland's rate fell by 62% in a single year, from 0.93 to 0.36.
- 40.1 million working days were lost to work-related ill health and injury in 2024/25. Injury accounted for 4.4 million of them; ill health for 35.7 million.
The national picture
Two systems measure workplace injury in Britain, and they measure different things. RIDDOR counts what employers are legally required to report — deaths, specified injuries, and injuries causing more than seven days off work. The Labour Force Survey asks workers directly, and therefore captures the injuries that never reach a report.
| Measure | Figure | Period | Source |
|---|---|---|---|
| Workers killed in work-related accidents | 126 | 2025/26 (provisional) | RIDDOR |
| Members of the public killed | 104 | 2025/26 (provisional) | RIDDOR |
| Fatality rate per 100,000 workers | 0.37 | 2025/26 (provisional) | RIDDOR |
| Workers injured at work | 680,000 | 2024/25 | Labour Force Survey |
| Injuries reported by employers | 59,219 | 2024/25 | RIDDOR |
| Injury rate per 100,000 workers | 2,070 | 2024/25 | Labour Force Survey |
| Reported injury rate per 100,000 employees | 209 | 2024/25 | RIDDOR |
| Working days lost to injury | 4.4 million | 2024/25 | Labour Force Survey |
| Average days off per injured worker | 6.5 | 2024/25 | Labour Force Survey |
HSE's published RIDDOR figures exclude incidents on railways, offshore installations, and workplaces enforced by the Office for Nuclear Regulation.
The reporting gap is the most important number here
59,219 employer reports against 680,000 self-reported injuries is a ratio of roughly one to eleven. Part of that gap is definitional: RIDDOR only requires the more serious injuries to be reported, so a strain that keeps you off for two days is not reportable in the first place.
But not all of it is definitional. HSE states that employer reporting of RIDDOR-defined non-fatal injuries is estimated at around half of what should be reported. In other words, even within the narrow category that legally must be reported, roughly one in two is not. The published figures are a floor, not a count.
This matters well beyond statistics. If you were injured at work and no RIDDOR report was made, you are in the overwhelming majority — and that is not a bar to claiming. What matters for a claim is what evidence exists, which is why accident book entries and medical records carry the weight they do.
How workers are killed: fatal injuries by cause
| Kind of accident | Deaths | Share of total |
|---|---|---|
| Falls from a height | 31 | 24.6% |
| Struck by a moving vehicle | 24 | 19.0% |
| Struck by a moving object | 21 | 16.7% |
| Trapped by something collapsing or overturning | 18 | 14.3% |
| Contact with moving machinery | 10 | 7.9% |
The five accident kinds shown account for 104 of the 126 deaths. RIDDOR 2025/26 provisional, HSE.
Falls from a height remain the single largest cause of workplace death, and have been for years. Read alongside the industry table below, that points at a specific and avoidable pattern: work at height in construction, agriculture and maintenance. HSE's own enforcement focus reflects it. If you have been injured in a fall at work, the fall from height guide sets out how liability and evidence work in those claims.
Where workers are killed: fatal injuries by industry
| Industry | Deaths |
|---|---|
| Construction | 25 |
| Agriculture, forestry and fishing | 22 |
| Manufacturing | 18 |
| Transportation and storage | 15 |
| Wholesale, retail and motor repair; accommodation and food | 11 |
| Administrative and support services | 9 |
| Waste and recycling | 6 |
| Other industries | 19 |
| Unknown | 1 |
RIDDOR 2025/26 provisional, HSE. Figures sum to 126.
Agriculture, forestry and fishing is the outlier in this table. Twenty-two deaths from one of the smallest workforces in the country produces a fatality rate several times the national average — the sector's risk is concentrated in lone working, machinery, and livestock handling. Construction's 25 deaths come from a workforce many times larger, and manufacturing's 18 and transportation's 15 follow the same pattern of machinery, vehicles and falls.
How workers are injured: non-fatal injuries by cause
The injury profile looks quite different from the fatality profile. These are the accident kinds accounting for 5% or more of employee injuries reported under RIDDOR in 2024/25:
| Kind of accident | Share of reported injuries |
|---|---|
| Slips, trips or falls on the same level | 30% |
| Handling, lifting or carrying | 17% |
| Struck by a moving object | 10% |
| Acts of violence | 10% |
| Falls from a height | 8% |
Accident kinds accounting for 5% or more of employee injuries reported under RIDDOR in 2024/25. HSE.
Slips, trips and falls on the same level are the most common injury — and the least feared. Handling, lifting and carrying at 17% is the second, and is the injury type most closely tied to a specific legal duty: the Manual Handling Operations Regulations 1992. Where an employer has not assessed a lift, the breach is often straightforward to establish. Manual handling claims turn on that assessment.
Injury rates by industry
Deaths tell you where the worst outcomes are. Rates tell you where injuries are most likely. Measured per 100,000 workers in the last 12 months, the ordering is genuinely different:
| Industry | Injuries per 100,000 workers |
|---|---|
| Accommodation and food services | 3,080 |
| Construction | 2,500 |
| Transportation and storage | 2,430 |
| Wholesale and retail trade | 2,250 |
| Administrative and support services | 2,160 |
| Manufacturing | 2,110 |
| Human health and social work | 1,860 |
| Education | 1,820 |
| Public administration and defence | 1,580 |
| Professional and scientific activities | 700 |
| All industries (average) | 1,780 |
Labour Force Survey self-reports, latest available year (2024/25). HSE flags accommodation and food services, construction, transportation and storage, and wholesale and retail as statistically significantly above the all-industry average.
Three things stand out. Accommodation and food services has the highest injury rate in the country at 3,080 per 100,000 — four times the rate in professional and scientific activities — yet it records no meaningful number of fatalities. Construction appears near the top of both lists, which is unusual. And professional, scientific and technical activities sits at 700, well under half the national average of 1,780.
The practical implication: the risk of being injured at work and the risk of being killed at work are largely different problems, in different sectors, needing different interventions. Reporting systems that track only deaths will systematically miss where most harm actually happens.
Regional and national variation
Fatality rates vary more between UK nations and regions than most people expect. The table below shows worker deaths and the rate per 100,000 workers for the latest year, the revised previous year, and the five-year total from 2021/22.
| Area | 2025/26 deaths | 2025/26 rate | 2024/25 deaths | 2024/25 rate | 5-year deaths | 5-year rate |
|---|---|---|---|---|---|---|
| East | 16 | 0.52 | 13 | 0.43 | 63 | 0.42 |
| North West | 15 | 0.41 | 21 | 0.59 | 87 | 0.49 |
| South East | 14 | 0.29 | 14 | 0.30 | 72 | 0.31 |
| South West | 14 | 0.46 | 10 | 0.34 | 70 | 0.47 |
| East Midlands | 12 | 0.51 | 7 | 0.30 | 48 | 0.41 |
| West Midlands | 11 | 0.38 | 6 | 0.21 | 48 | 0.33 |
| Yorkshire and The Humber | 9 | 0.33 | 6 | 0.22 | 44 | 0.33 |
| London | 7 | 0.13 | 11 | 0.20 | 56 | 0.21 |
| North East | 5 | 0.41 | 3 | 0.25 | 15 | 0.25 |
| Wales | 9 | 0.62 | 10 | 0.69 | — | — |
| Scotland | 10 | 0.36 | 26 | 0.93 | — | — |
| Great Britain | 126 | 0.37 | 128 | 0.38 | — | — |
RIDDOR, HSE. 2025/26 figures are provisional; 2024/25 figures are revised. Rates are per 100,000 workers. Five-year figures cover 2021/22 to 2025/26 and are published by HSE for England's regions only.
Wales recorded the highest rate in Great Britain at 0.62 deaths per 100,000 workers, followed by the East and the East Midlands at 0.52 and 0.51. London recorded the lowest at 0.13 — under a quarter of the Welsh rate. The North West recorded the highest five-year total of any English region at 87 deaths.
The sharpest movement is Scotland, where worker deaths fell from 26 to 10 and the rate from 0.93 to 0.36 — a 62% reduction in a single year. A swing of that size in a comparatively small number of incidents should be read with care: year-on-year regional counts are volatile, which is exactly why the five-year column is included. Scotland's 2024/25 figure was itself well above its longer-run average.
For anyone injured in one of these regions, the local picture is not what determines a claim — the type of accident and the evidence available do. But the regional spread is a useful corrective to the idea that workplace safety is a uniform national problem.
Working days lost
40.1 million working days were lost in 2024/25 to work-related ill health and injury. Of that, 4.4 million days were lost to non-fatal injury and 35.7 million to ill health. Each person injured took an average of 6.5 days off; each person with work-related ill health took 19.1 days.
Those figures cross-check cleanly: 680,000 injured workers at an average of 6.5 days each is 4.42 million days, matching HSE's published 4.4 million. It is worth noting what sits beside it — stress, depression and anxiety alone accounted for 22.1 million days lost, and musculoskeletal disorders 7.1 million. Ill health, not injury, is where the overwhelming majority of working time is lost.
What this means if you have been injured at work
- Most injuries are never reported, and that does not stop a claim. With roughly one RIDDOR report for every eleven injuries, an unreported injury is the normal case rather than an obstacle. Evidence can be assembled after the event.
- Your industry affects the legal duties, not your right to claim. Construction, manufacturing and agriculture carry specific regulations — CDM 2015, PUWER, the Provision and Use of Work Equipment Regulations, and the Work at Height Regulations 2005 — that make breaches easier to establish than a general negligence argument.
- Time is the real constraint. Most claims must be brought within three years, and medical evidence takes time to obtain. The time limit guide explains the exceptions.
How to cite this page
Every figure here is HSE's. The analysis — the reporting ratio, the fatal versus non-fatal industry comparison, and the regional commentary — is ours, and is free to use with attribution. Suggested citation:
WorkClaim UK, Workplace Injury Statistics in Great Britain, 8 October 2026. Based on Health and Safety Executive data. https://accident-at-work-claims.com/workplace-injury-statistics
Journalists and trade bodies wanting comment, a regional breakdown, or the underlying figures in spreadsheet form can contact us here. WorkClaim UK is operated by Edward and Amaury Limited (Edward & Amaury Solicitors), SRA number 800525.
Methodology and sources
- Fatal injuries (2025/26). RIDDOR reports of work-related fatal injuries to workers, published by HSE as provisional and to be finalised in July 2027. Includes employees and self-employed people. Excludes deaths to patients and service users in healthcare and adult social care in England.
- Non-fatal injuries and working days lost (2024/25). HSE's Labour Force Survey estimates and employer reports under RIDDOR. RIDDOR non-fatal figures exclude incidents on railways, offshore, and in premises enforced by the Office for Nuclear Regulation.
- Industry and regional rates. HSE's RIDFATAL dataset for fatalities, and Labour Force Survey self-reports for non-fatal injury rates. Rates are per 100,000 workers. Survey-based rates carry a sampling confidence interval, and HSE flags the sectors that differ statistically significantly from the average.
- Comparability caution. Because reporting levels differ between sectors, HSE advises care when comparing RIDDOR subsets with one another. That is why the industry injury rates above use the survey-based measure rather than RIDDOR counts.
Primary sources: HSE — Work-related fatal injuries in Great Britain, HSE — Non-fatal injuries at work in Great Britain, HSE — Working days lost, and HSE — Industries. All HSE statistics are published under the Open Government Licence v3.0.
Frequently asked questions
How many people are killed at work in Great Britain each year?
126 workers were killed in work-related accidents in 2025/26, according to provisional RIDDOR figures published by the Health and Safety Executive. A further 104 members of the public died in work-related accidents. The rate was 0.37 deaths per 100,000 workers, and HSE describes the long-term trend as downward.
How many workplace injuries go unreported?
The great majority. HSE's Labour Force Survey estimates 680,000 workers sustained a non-fatal injury at work in 2024/25, while employers reported 59,219 injuries under RIDDOR — about 8.7%. HSE separately estimates that employer reporting covers only around half of the injuries that are actually reportable, so the shortfall has two causes: injuries below the reporting threshold, and under-reporting of those above it.
What is the most common kind of workplace accident?
Slips, trips and falls on the same level accounted for 30% of employee injuries reported under RIDDOR in 2024/25, followed by handling, lifting and carrying at 17%. But the most common fatal accident is different: falls from a height caused 31 of the 126 worker deaths in 2025/26 — nearly a quarter of the total.
Which industry is the most dangerous to work in?
It depends whether you measure deaths or injuries, and the answers differ. Construction recorded the most worker deaths in 2025/26 (25), followed by agriculture, forestry and fishing (22) — which is far more dangerous per worker given how small that workforce is. On non-fatal injury rate, accommodation and food services is highest at 3,080 per 100,000 workers, ahead of construction at 2,500.
Does my injury need to be in the accident book to make a claim?
No. RIDDOR reporting only covers the more serious injuries, and HSE estimates that even reportable injuries are substantially under-reported by employers. Most workplace injuries never generate a RIDDOR record, and many are not entered in the accident book either. A claim can still be proved through medical records, witnesses, photographs and other evidence — but an early accident book entry or RIDDOR report makes the case considerably easier.
Where does this data come from?
Every figure on this page is published by the Health and Safety Executive. Fatality figures are RIDDOR reports for 2025/26 and are provisional until finalised in July 2027. Non-fatal injury, working days lost and industry-rate figures are HSE's 2024/25 releases, combining employer reports under RIDDOR with self-reports from the Labour Force Survey.