Quiet hazards at work are becoming harder to ignore

The Lloyd’s Register Foundation World Risk Poll has become one of the few genuinely global sources that tells us not only what people fear, but what they say they have actually experienced. That matters for OSH. In many parts of the world, official data on workplace harm are still incomplete, especially for workers in informal, precarious or hard-to-reach jobs. The World Risk Poll helps make these risks more visible. 

In 2025, 16% of the global workforce reported personally experiencing serious harm at work in the past two years. That is down from 18% in 2023 and 19% in 2021. This is encouraging. It suggests that prevention, regulation, training, reporting systems and worker participation can make a difference.

But 16% is still a very high number: around 610 million workers harmed on the job. Workplace harm remains a common, systemic risk. The report also shows that harm is unevenly distributed. Workers with lower education and lower financial resilience face higher risks, and country and regional differences remain large. This matters for policy. A global average can show direction, but national action must be built around the workers, sectors and workplaces where risk is concentrated.

Concern is not the problem

One of the most important patterns in the published World Risk Poll data is the gap between worry and harm. Between 2021 and 2025, global worry about harm from work rose from 48 per cent to 51 per cent, while reported harm fell. In simple terms, the ratio between worry and reported harm is increasing. Using the Poll series, global worry-to-harm for work rises from roughly 2.5 in 2021, to just over 3 in 2025.

This should not be read as bad news. It may mean that more workers no longer accept harm at work as normal. Awareness and concern can create pressure for change. They can help workers demand safer systems, encourage employers to act earlier, and push governments to make the ILO’s fundamental principle and right at work – a safe and healthy working environment – a reality.

What does “harm” capture?

Here we need to be careful. The World Risk Poll asks workers whether they have experienced “serious harm” from work. In practice, many respondents may think first of physical injuries and accidents. That is also where official OSH statistics are strongest. International statistical standards exist for occupational injuries, and many national systems collect accident and injury data more regularly and comprehensively than data on occupational diseases, hazardous exposures or psychosocial risks.

This alignment is useful. The Poll’s finding of declining workplace harm broadly mirrors trends seen in many official occupational injury statistics, suggesting that respondents may often interpret “harm” primarily as injury. But it also exposes an important gap. The ILO’s concept of a safe and healthy working environment is much broader than the prevention of traumatic injuries alone. It includes occupational diseases, mental health, psychosocial risks, violence and harassment, excessive working time, exposure to hazardous substances, excessive heat, and many other factors that can damage health gradually and often invisibly.

The 2025 World Risk Poll itself points in this direction. In high-income countries, worry about workplace harm rises sharply among workers who work 50 or more hours per week, even where reported harm is not higher. This suggests that many workers increasingly associate occupational risk not only with physical injury, but also with long hours, psychological strain, stress and burnout. 

The broader evidence is even more striking. ILO estimates for 2019 put the global number of work-related deaths at approximately 2.93 million. Of these, around 2.6 million were caused by occupational diseases, while about 330,000 resulted from fatal occupational accidents. In other words, the vast majority of work-related deaths are linked to disease rather than injury.

The ILO is also drawing growing attention to psychosocial risks and mental health at work. Poor mental health is estimated to result in the loss of around 12 billion working days every year, while psychosocial risks at work are increasingly recognized as a major occupational health challenge. Recent evidence suggests that psychosocial risks, including working more than 55 hours a week, may be linked to more than 840,000 deaths annually. If we focus only on visible physical injuries, we risk overlooking much of the true burden of work-related harm.

Evidence must catch up with the duty of prevention and right to safety and health

The World Risk Poll is valuable because it captures workers’ experiences directly, including in places where official systems remain weak. But it should be seen as a signal, not a substitute. The goal is not simply to know that workers are worried. The goal is to turn that concern into prevention and to make the fundamental right to a safe and healthy working environment a reality in practice.

The practical lesson is straightforward. We need statistics that reflect the full reality of work. That means strengthening data on occupational injuries while also improving the measurement of occupational diseases, harmful exposures, mental health and psychosocial risks. It means combining administrative records with worker surveys and other data sources, so that underreporting does not leave the most vulnerable workers invisible. And it means using this evidence to guide labour inspection, prevention programmes, training, regulation and social dialogue.

This is where the 22nd International Conference of Labour Statisticians (ICLS) becomes important. The current international statistical framework on OSH remains largely centred on occupational injuries and accidents, while major gaps persist in the measurement of occupational diseases, psychosocial risks, hazardous exposures and broader indicators of OSH system performance. The direction of travel is clear: the 22nd ICLS in 2028 is expected to consider updated standards that move OSH statistics from fragmented and reactive reporting towards integrated, prevention-oriented systems.

For OSH, the next step is to connect three things: what workers experience, what workers worry about, and what official systems measure. Worry is not a substitute for evidence, but it is evidence of its own kind. When workers worry more, even without recent injury, policymakers should listen. It may be the early warning signal needed to strengthen prevention before harm occurs. 

The quiet hazards are no longer quiet.