The Risk Of RIDDOR – Are You Counting Outcomes Instead Of Warnings?

For senior leaders, boards, safety professionals and operational managers, the question is not whether RIDDOR (the Reporting of Reporting of Injuries, Diseases and Dangerous Occurrences Regulations) matters. It does. The important question is whether it has become a substitute for understanding how well risks are actually being controlled.

Imagine two organisations.

The first recorded three RIDDOR-reportable incidents last year.

The second recorded none.

Which is safer?

At first glance, the answer appears obvious. Three reportable incidents suggest failure. Zero suggests control.

One organisation appears to have a problem; the other appears to be performing well.

But now add some context.

The first organisation actively encourages reporting. Workers raise concerns without fear of blame. Near misses are discussed. Supervisors investigate events properly. Corrective actions are tracked. Senior managers ask difficult questions. Lessons are shared across departments, and uncomfortable findings are not quietly edited out of reports.

The second organisation has a very different culture.

• Minor incidents are dismissed as “part of the job”
• Near misses are rarely recorded
• Supervisors discourage paperwork
• Workers believe that reporting a problem will make them look incompetent
• Defects remain in use because production is busy
• Investigations stop at “operator error”

Therefore, senior managers receive a monthly dashboard showing another reassuring green box stating “zero RIDDOR incidents”.

So, which organisation is safer now?

The answer is no longer quite so simple. And that is the problem.

Throughout this article, we’ll explore exactly why a RIDDOR-only mindset isn’t sufficient for modern organisations. But first, let’s touch on why we’ve chosen now to shine a light on this topic.

A timely question – why is RIDDOR back under scrutiny?

In April 2026, the Health and Safety Executive launched a public consultation on the Reporting of Injuries, Diseases and Dangerous Occurrences Regulations 2013. The proposals include clarifying definitions, revising the lists of dangerous occurrences and occupational diseases, broadening who may diagnose reportable occupational disease, and simplifying the online reporting process to reduce both under-reporting and over-reporting.

That review matters.

RIDDOR provides an essential statutory reporting framework. Serious work-related events must not disappear into internal filing systems. Regulators need information to identify significant risks, analyse patterns and direct attention where it is needed.

But the current discussion creates an opportunity to ask a broader question:

Have some organisations allowed a legal reporting threshold to become a measure of safety performance?

Because legal reporting mechanisms and safety performance are not the same thing.

The distinction is consistent with established health and safety management thinking: organisations need both incident-led information and routine, preventive monitoring. Boards should receive evidence about what has gone wrong, but also evidence about whether training, maintenance, supervision, risk controls and worker engagement are functioning before harm occurs.

RIDDOR tells us that a defined reportable event has occurred. It does not, by itself, tell us whether an organisation has a strong safety culture. It does not tell us how many warnings came first. And it certainly does not prove that an organisation with zero reports is free from serious risk.

The comfort of zero – RIDDOR’s biggest risk

There is something psychologically powerful about zero.

• Zero accidents
• Zero lost-time injuries
• Zero RIDDOR reports
• Zero enforcement action

It looks clean on a dashboard. It fits neatly into a board report. It can be presented in green and accompanied by an upward arrow labelled “improvement”.

But zero can mean several things. It may mean that an organisation has genuinely effective controls, competent supervision, strong worker engagement and a mature learning culture. It may also mean that nobody reported anything.

A genuinely safe workplace and a workplace with a poor safety culture can look identical on a spreadsheet.

This is where safety performance measurement becomes dangerous. Not because accident data is useless—it is not—but because an absence of recorded outcomes can be mistaken for evidence that risk is being effectively controlled.

A quiet reporting system is not necessarily a safe system.

Sometimes it is simply quiet.

The rear view mirror problem

Most organisations would never attempt to drive a vehicle by looking only in the rear-view mirror. Yet some manage safety in almost exactly that way. They review what has already happened:

• How many accidents?
• How many lost-time injuries?
• How many RIDDOR reports?
• How many days since the last incident?
• How many claims?

These measures matter. They show consequences. They reveal trends. They can identify recurring failures. But they are fundamentally retrospective. By the time a serious injury appears on the dashboard, the control system has already failed.

• The worker has already been hurt
• The investigation has already started
• The damage has already been done

The question is not simply:

“What happened last month?”

It is:

“What is happening in our workplace today that could become tomorrow’s incident?”

Before the accident, there is often a story

Serious incidents rarely arrive without context:

• Before the injury, there may have been a damaged guard.
• Before the fall, there may have been repeated shortcuts.
• Before the vehicle collision, there may have been poor segregation.
• Before the exposure, there may have been ineffective LEV.
• Before the electrical incident, there may have been a temporary repair that became permanent.
• Before the manual handling injury, there may have been months of complaints.
• Before the machinery accident, someone may already have said: “We’ve been doing it this way for years.”

These are warning signs.

Individually, they may appear insignificant. Collectively, they reveal a system drifting towards failure. The problem is that many organisations do not measure them. They measure the injury. They do not measure the deterioration that came before it.

What are you actually counting?

Consider a typical monthly safety report. It may include:

• number of accidents
• number of lost-time injuries
• number of RIDDOR reports
• number of days without injury
• number of enforcement notices

Now consider what may be absent.

A more useful report would also show whether the organisation is finding and fixing early warning signs, such as:

• overdue safety-critical defects
• repeat corrective actions
• risk assessments that no longer reflect actual work
• temporary controls that have become normal practice
• worker concerns
• deferred maintenance
• serious-potential near misses
• weak critical controls.

Asking questions about these pre-cursors to accidents and incidents is much harder. It’s also more useful.

What to measure instead

A stronger dashboard should balance lagging outcomes with measures that test whether the organisation is controlling risk in real time. Useful indicators may include the percentage of safety-critical actions closed on time, the age of high-risk defects, the quality of investigations, verification of critical controls, repeat findings, worker-raised concerns, serious-potential near misses, maintenance backlog, exposure monitoring results and evidence that lessons have been shared and embedded.

The point is not to create more numbers for their own sake. It is to choose measures that prompt action: what needs attention, who owns it, when will it be fixed, and how will the organisation know the control is effective?

Not all near misses are equal

There is another trap. Some organisations discover the concept of leading indicators and immediately begin counting near misses. The target becomes:

“We need 50 near-miss reports this month.”

The result?

A sudden increase in reports about empty boxes, minor housekeeping issues and low-consequence observations.

The target has been achieved. The dashboard turns green.

But has risk reduced?

Not necessarily. The value of a near miss is not simply that it was counted. The value lies in what it reveals. Consider the below questions:

• Is a forklift narrowly missing a pedestrian equivalent to finding a coffee cup in the wrong waste bin?
• Is a dropped object from height equivalent to a slightly untidy storeroom?
• Is an unexpected machine movement during maintenance equivalent to a missing floor marking?

Clearly not. All deserve proportionate attention, but they do not carry the same potential consequence. A mature safety system does not simply count events. It understands risk significance. It asks:

What could realistically have happened?

That question may often tell you far more than the outcome.

When good numbers create bad behaviour

Safety targets can influence behaviour in unexpected ways. Tell a workforce that the objective is “zero accidents” and the intention sounds positive. But what happens when a team has worked 300 days without an accident and someone suffers a minor injury?

• Will they report it? Or will they feel responsible for “breaking the record”?
• What happens when a manager’s performance rating is linked to injury numbers?
• What happens when a contractor believes reporting an incident could affect future work?
• What happens when a supervisor knows that another recorded accident will trigger senior management scrutiny?

The metric intended to improve safety can begin to suppress the information needed to manage it. This is one of the uncomfortable realities of safety culture:

People respond to what organisations reward, punish and pay attention to.

If reporting bad news creates pain, eventually the bad news stops travelling upwards. The risk does not disappear with it.

RIDDOR is a legal requirement, not a culture score

RIDDOR has an important and specific function.

It requires defined work-related deaths, injuries, occupational diseases and dangerous occurrences to be reported by the appropriate responsible person where the legal criteria are met. That matters.

But RIDDOR was not designed to answer every question about organisational safety performance. A company can comply with RIDDOR and still have weak supervision. Here’s why:

• Organisations can submit every required report and still tolerate poor standards.
• Organisations can have no reportable incidents and still operate with serious uncontrolled risks.
• Organisations can have several reportable incidents and, in response, develop a stronger learning culture.

Number alone rarely tell the whole story.

Context matters. Reporting culture matters. Exposure matters. Risk profile matters. Learning matters. Leadership matters.

The difference between lagging and leading

Traditional accident statistics are often described as lagging indicators. They show outcomes. They show that something has already happened.

Leading indicators attempt to look at what’s happening earlier in the chain. They examine activities, conditions and control systems that may influence future performance.

Examples may include:

• completion of safety-critical actions
• verification of machine guarding
• quality of workplace inspections
• closure time for high-risk defects
• effectiveness of training
• worker participation
• supervisor engagement
• preventive maintenance performance
• exposure monitoring
• health surveillance trends
• permit-to-work compliance
• verification of critical controls

But there is an important warning here. A leading indicator is not useful simply because it is proactive.

• Counting 100 inspections tells you very little if the inspections are poor
• Recording 200 safety conversations tells you very little if they are superficial
• Achieving 100% training completion tells you very little if workers cannot apply the training
• Closing every corrective action tells you very little if the actions did not address root causes

The question is not:

“Did we do it?”

The better question is:

“Did it work?”

The green dashboard illusion

There is a particular kind of danger in a dashboard where everything is green:

• All inspections completed.
• All training complete.
• No overdue actions.
• Zero RIDDORs.
• Zero lost-time injuries.

Everything appears controlled. But mature organisations should be cautious when performance information becomes too comfortable.

A healthy reporting culture often produces findings.

• It identifies defects.
• It exposes weak controls.
• It generates corrective actions.
• It reveals disagreement.
• It uncovers uncertainty.

In other words, a strong safety culture may sometimes look messier than a weak one. That is not failure. It may be evidence that people are looking.

The question for senior leaders is, therefore, not whether the dashboard contains red. It is whether the organisation understands why it is red and what is being done about it. And if the dashboard is green, the question is why is it green?

What should senior leaders ask?

Perhaps the monthly board discussion should move beyond:

“How many accidents did we have?”

And begin asking:

• “Which of our critical controls are showing signs of weakness?”
• “What are workers telling us that we are not yet acting on?”
• “Which corrective actions keep recurring?”
• “Where are temporary controls becoming normal practice?”
• “Which near misses had serious potential?”
• “What are supervisors seeing at the point of work?”
• “Which risk assessments no longer match reality?”
• “Where are production pressures influencing behaviour?”
• “What bad news are we not hearing?”

That final question may be the most important. Because organisations rarely fail solely because hazards existed. They often fail because information about those hazards did not reach the people capable of acting on it. Or it reached them and was not treated with sufficient urgency.

From reporting culture to learning culture

Encouraging more reporting is useful. But reporting alone is not enough. Here’s why:

• If workers repeatedly raise concerns and nothing changes, reporting will eventually decline.
• If near misses disappear into a database, engagement will decline.
• If investigations blame individuals without examining systems, trust will decline.
• If corrective actions remain overdue, credibility will decline.

A reporting culture becomes a learning culture only when information leads to visible action. The loop must close.

Report: make it easy and safe for workers, supervisors and contractors to raise concerns, including weak signals and serious potential near misses.

Understand: examine underlying causes, risk significance, repeated patterns and organisational factors rather than stopping at individual behaviour.

Act.

Verify.

Share.

Without that final discipline, organisations risk becoming very good at collecting safety data while remaining poor at learning from it.

The KPI that changes the conversation

Instead of asking only:

“How many accidents did we have?”

Try asking:

“How many opportunities did we have to prevent an accident?”

That question changes the conversation. It makes a defect valuable information. It makes a near miss an opportunity to learn. It makes a worker concern a source of intelligence. It makes an overdue action visible. It makes supervision proactive. It moves safety away from counting injuries and towards understanding risk. And perhaps most importantly, it reinforces a simple cultural message:

You do not have to wait for someone to get hurt before you are allowed to improve something.

The insight challenge

Your organisation has recorded zero RIDDOR-reportable incidents this year. Good news. Now remove that figure from the dashboard and ask these questions:

• What other evidence can you produce that proves your organisation is becoming safer?
• Can you demonstrate stronger critical controls?
• Faster closure of serious defects?
• Better worker engagement?
• Improved supervision?
• More effective learning?
• Reduced exposure?
• Better maintenance reliability?
• Higher-quality investigations?
• Earlier identification of weak signals?

If your answer to any of these questions is no, then perhaps zero is telling you less than you think.

Final thoughts on safety culture:

RIDDOR counts serious defined outcomes that meet legal reporting criteria. Organisations need to count something else as well

• They need to count the warnings.
• The weak signals.
• The repeated defects.
• The overdue actions.
• The serious-potential near misses.
• The controls that are beginning to drift.
• The concerns raised by workers.
• The lessons that have not yet been learned.
• Because the safest organisations are not simply those with the fewest reports.
• They are the organisations that remain curious when the numbers look good.
• They do not confuse silence with safety.
• They do not wait for an injury to prove that a control was weak.

In short, the best organisations understand one of the most important principles in modern safety management:

RIDDOR tells you where you have been. Your culture, controls and capacity to learn will influence where you go next.

Remember…

The absence of an accident is not proof of the presence of safety. A mature organisation does not simply ask whether anyone was hurt. It asks what nearly happened, what is changing, what is deteriorating and whether the warning signs are being heard before they become statistics.

For more information about this article or any other aspect of Health and Safety or Employment Law, download our App for Apple or Android.