What Are We Not Seeing?

Safety professionals spend a lot of time talking about injury rates.

Injury statistics tell us what happened. Understanding risk requires us to ask what we might be missing.

LTIFR, TRIFR and other injury frequency measures appear in board reports, tender submissions, monthly reviews and annual reports. They provide useful information and can help identify trends over time. The problem is not that these measures exist. The problem is the weight we often give them.

If an organisation’s injury rates begin to climb, people pay attention. Questions are asked. Investigations are launched. Leaders want to understand what has changed and what needs to be done differently.

When injury rates are low, the opposite can occur. There is often a sense that things are going well. Dashboards are green. Targets are being achieved. Reports look positive. None of this is inherently bad, but it can create an assumption that risk is being effectively managed simply because injuries are not occurring.

History tells us that this assumption can be dangerous.

Many serious incidents and fatalities occur in organisations that have otherwise strong injury statistics. They happen while performance indicators appear healthy, targets are being met, and people genuinely believe things are under control.

This is where injury rates reach their limit.

Like any metric, injury rates can influence behaviour. The more attention we place on a number, the greater the temptation to focus on the number itself.

In some cases, the metric can become the goal.

“When the metric becomes the goal, we stop managing risk and start managing numbers.”

When that happens, conversations shift from understanding risk to managing statistics. Organisations can become highly effective at improving injury rates while learning very little about whether risk is actually being reduced.

Concerns about relying too heavily on injury frequency rates are not new. In 2025, Safe Work Australia removed its LTIFR calculator, recognising the limitations of using injury frequency rates in isolation to measure WHS performance.

An injury statistic can tell us what happened last month. It cannot tell us whether critical controls are working today. It cannot tell us whether workers have stopped raising concerns. It cannot tell us whether a risky workaround has quietly become standard practice. Most importantly, it cannot tell us what might happen tomorrow.

In that sense, injury rates are a little like a rear-view mirror. Looking in the mirror is useful. In fact, it would be reckless not to. But nobody drives safely by staring at the rear-view mirror alone. The real value comes from understanding what is ahead, what is changing, and what requires attention.

Perhaps that is why some of the most effective organisations focus less on the outcome itself and more on understanding the risks that create it.

Instead of asking only how many people were injured, they ask different questions.

What risks have we reduced?

What risks have we accepted?

What concerns are workers raising?

Which critical controls have we verified?

What are we learning that we did not know yesterday?

These questions are harder to answer than an injury rate. They are also far more revealing.

The organisations that consistently improve are rarely the ones focused solely on injury statistics. They are the ones still asking difficult questions when everything appears to be going well. They remain curious when others become comfortable.

So perhaps the question is not whether injury rates matter.