HSE Statistics: A UK Workplace Injury and Ill-Health Breakdown

The HSE statistics are the foundation document for any UK employer trying to understand workplace risk in their sector. They are published annually in late autumn — typically the third week of November — covering the financial year that ended the previous March. The 2024/25 figures, published November 2025, are the most recent at the time of writing. The 2025/26 figures will land in November 2026 and we’ll refresh this article then.

This guide breaks down what the figures actually show, how to read them sensibly, where the rolling trends are heading, and what employers should be doing with the data. It is written for the H&S manager, operations director or business owner who needs to translate the numbers into an operational position rather than just quote a headline statistic in a board paper.

The headline figures

Two notes on reading the figures. First, fatal injuries are recorded with high reliability — every workplace fatality reaches HSE one way or another. Reportable non-fatal injuries are recorded with moderate reliability — there is widely-acknowledged underreporting under RIDDOR, particularly of over-7-day injuries. Self-reported injuries from the Labour Force Survey are a better measure of what actually happens in workplaces because they don’t depend on the employer reporting accurately. Use the LFS figure when you want the population-level reality; use the RIDDOR figure when you want the regulatory reporting picture.

Second, the comparison with five years ago has been distorted by the pandemic. The 2020/21 figures are not comparable with normal years on injury (because activity dropped) or on ill health (because work-related stress definitions shifted with the move to remote working). Trend analysis that runs through 2020/21 needs that caveat applied.

Fatal injuries — what the figures show

The 124 worker fatalities in 2024/25 sit within the range of recent years (135 in 2023/24, 138 in 2022/23, 123 in 2021/22). The headline number conceals concentrations:

• Construction: ~35 fatalities. Highest of any sector. Falls from height remain the leading single cause within construction — close to half of all construction fatalities.
• Agriculture, forestry and fishing: ~23 fatalities. Highest fatality rate per 100,000 workers of any major sector by a considerable margin. Vehicle incidents and being struck by moving objects are the dominant causes.
• Manufacturing: ~11 fatalities. Slowly declining over the long run. Remaining fatalities concentrated in food manufacture and engineering.
• Transport and storage: ~12 fatalities. Workplace transport (forklifts, vehicle reversing, loading bay incidents) is the dominant pattern.
• Waste and recycling: ~6 fatalities. A small sector with a high fatality rate per worker; well above the national average.

The leading causes across all sectors, ranked: falls from height (~30 per year), being struck by a moving vehicle (~20 per year), being struck by a moving object (~15 per year), trapped by something collapsing or overturning (~10 per year), contact with moving machinery (~10 per year). These five categories account for around two-thirds of worker fatalities every year and have done for more than a decade.

Work-related ill health — the rising signal

The ill-health figures are where the biggest signal lives. Around 1.7 million workers reported work-related ill health in 2024/25 — broken down roughly as:

The stress signal deserves particular attention. Work-related stress, depression and anxiety has gone from being a recognised but secondary category 15 years ago to comfortably the largest category of work-related ill health today. The drivers are debated — pandemic effects, hybrid working transition, broader societal mental-health awareness, genuine increase in work intensity — but the operational implication for employers is clear. The HSE Management Standards approach to stress that we cover in the stress at work guide is no longer a niche topic; it is core H&S management.

Musculoskeletal disorders (MSDs) are the second-largest category, at around 543,000 reported cases. The MSD signal is slowly improving over the long run but has been flat in recent years, with the move to home and hybrid working showing up as a marginal increase in upper-limb and back disorders. The manual handling regulations and DSE regulations are the primary control framework, and the rise in DSE-related MSDs is the visible cost of inadequate hybrid-working DSE assessments.

The sectoral picture

Total fatalities across sectors give a broad shape, but the per-worker fatality rate gives a much sharper picture of relative risk. The latest figures, normalised per 100,000 workers:

The pattern is consistent year on year. Agriculture is uniquely dangerous; construction is the largest sector with the largest absolute count but improving; manufacturing is now broadly average for the economy as a whole. For an employer reviewing their own performance, the per-worker rate against the sector benchmark is a far more useful measure than the absolute number against last year — sector matters enormously.

What the figures don’t tell you

The HSE statistics are the best available data on UK workplace H&S performance, but they have known limitations the careful reader should know about:

1. RIDDOR underreporting

HSE itself estimates that around half of all RIDDOR-reportable over-7-day injuries are not actually reported. The RIDDOR figure is therefore a floor rather than a ceiling. The Labour Force Survey self-reported figure (~600,000 non-fatal injuries) is closer to the real number than the ~60,000 RIDDOR figure suggests.

2. Pandemic distortion in trend lines

2020/21 was an atypical year for everything — workplace activity, reporting, ill-health categorisation. Trend lines that include 2020/21 are noisy. The cleanest trend analyses run pre-2020 and post-2022 separately, and treat 2020/21 as an outlier.

3. The figures are workforce-level, not workplace-level

The statistics tell you how the UK economy as a whole is performing. They tell you nothing direct about how your specific workplace is performing. An employer with zero injuries this year cannot conclude their controls are working — sample size of one against a sector mean of 0.4/100,000 is statistically silent.

4. The Cost to Britain figure is an estimate, not a measurement

The ~£21.6 billion annual cost figure is calculated from injury and ill-health prevalence multiplied by economic-loss assumptions. It is useful for board-level conversations and policy work but should not be quoted as if it were a directly measured quantity.

5. Mental health categorisation has shifted

The boundary between work-related stress, depression and anxiety on the one hand and personal mental health conditions exacerbated by work on the other has moved over the last decade. Some of the apparent rise in work-related stress is a genuine increase; some is a category shift. The figures over a 10-year window need that caveat.

What employers should actually do with the data

Five practical uses for the HSE statistics in operational H&S management:

1. Benchmark your own performance against the sector mean. The per-worker fatality and injury rates by sector give you the comparator. If your incident rates sit well above the sector mean, the data is telling you something. If they sit well below, you may be underreporting.
2. Use the leading-cause data to prioritise your risk register. If falls from height account for half of construction fatalities, the construction firm whose risk register doesn’t have working at height as the top item is misaligned with where the harm is actually happening. The same logic applies sector by sector.
3. Bring the rising-stress signal into your management agenda. The stress-at-work figures have been the strongest signal in the data for the last decade and the slowest one for most employers to act on. Boards that see the figures and respond with “we’ve covered mental health in the wellbeing programme” are likely under-controlled.
4. Use the Cost to Britain figure for executive-level conversations. Some senior leaders do not engage with H&S as a moral issue but do engage with the financial scale. The £21.6 billion annual cost figure is the right framing for those conversations.
5. Track year-on-year changes alongside the HSE figures, not in isolation. Your own incident data tracked against the sector trend tells you whether you are improving faster or slower than the industry. Improving in absolute terms while the industry improves faster means you are losing ground.

For training, the HSE statistics should inform the curriculum at every level. IOSH Working Safely at employee level uses the figures to anchor risk awareness; IOSH Managing Safely at supervisor level uses them to shape risk assessment priorities; the NEBOSH National General Certificate at H&S manager level treats the statistics as a foundational reference document. Where an organisation needs help making sense of its own figures against the sector benchmarks, our consultancy team can scope a structured performance review and KPI alignment exercise. Call us on +44 (0) 3300 569534 for tailored advice.