You bought the EAP. So why aren't your people using it?

Here is the conversation, almost word for word, in roughly half the operations reviews we sit in. We pay for an EAP every year. The utilisation report comes back in single digits, again. And psychological claims keep climbing. Leadership has said mental health is a priority — it is in the strategy deck — but the numbers say nothing on the ground has changed. So what are we actually paying for?

It is a fair question, and the honest answer is uncomfortable: you are paying for a backstop and treating it as a strategy. An EAP is reactive support that activates after harm has already landed. Low utilisation is not the failure. It is the system working exactly as designed — and the design has a hole in it.

Why doesn't anyone use the number on the poster?

Start with the scale of the gap. Roughly 80% of Australia's top 500 companies offer an EAP, yet average utilisation sits at about 5.18% (Sonder, 2025). That is not a rounding error. That is the overwhelming majority of your workforce, every year, not touching the one mental-health control most organisations can name.

The instinct is to read that as an awareness or stigma problem and respond with another comms campaign. But the comms are not the constraint. The Corporate Mental Health Alliance Australia found that 64% of employees say their organisation values mental health, while only 51% feel safe raising a mental-health concern at work (CMHAA, 2024). People know the EAP exists. They have done the maths on what it costs to use it — visibility, perceived weakness, the fear it travels — and most decide the threshold for picking up the phone is "things have gone badly wrong". By then you are managing a crisis, not preventing one.

So the EAP captures the tail: the small fraction already in acute distress. Everyone in the long, slow slide before that point — the manager carrying a team through restructure, the supervisor absorbing every complaint on a site, the high performer quietly going under — sits in a space no control touches. That is where most people-risk lives, and right now nobody in your operating model owns it.

What is the EAP actually a control for?

This is where the framing matters more than the spend. Treat psychosocial risk the way your business already treats physical risk and the EAP's place becomes obvious. You would never run a site on first aid alone. First aid is what you reach for after someone is hurt. It does not appear at the top of any hierarchy of controls, because the point of the hierarchy is to eliminate or reduce the hazard before it can do harm.

An EAP is psychological first aid. It is the bottom of the cliff. Useful — necessary, even — but it is the last line, not the system. Under the model WHS regulations, psychosocial hazards are now explicitly something a person conducting a business must identify, assess and control so far as is reasonably practicable. That is a duty, and it sits with officers as part of due diligence. A utilisation report showing 5% is not evidence that you have discharged it. It is closer to evidence that your only control sits at the very bottom of the hierarchy, where it is least effective.

The regulator is not asking whether you offer support after harm. It is asking what you are doing to make the invisible visible and reduce exposure before harm occurs. Most organisations cannot answer that, because the layer that would answer it — the pre-EAP layer — has never been built.

What does the gap cost while it stays invisible?

The line that should focus the room is this: psychological claims carry the longest time lost and the highest cost of any claim type, with affected workers off for around 34 to 35.7 weeks (Safe Work Australia, cited in SafetySure, 2025). Compare that to a typical physical injury and the asymmetry is stark. One psychological claim is not one claim's worth of cost — in lost time, replacement, premium impact and management drag, it is several.

And these are the claims your EAP, by its position in the system, is structurally worst at preventing. It only meets people once they have crossed into distress. The eight or nine months of declining performance, withdrawal and rising risk that precede a claim happen entirely in the gap. The cost is already accruing on your P&L; it is simply not labelled yet.

There is a leadership layer to this too. The CMHAA survey found 84% of organisations say mental health is a priority, but only 52% of leaders feel confident actually addressing it (CMHAA, 2024). That gap — stated priority, absent capability — is exactly what produces an unused EAP and a rising claims line at the same time. Your managers are sitting at Dunbar's number, close enough to 150 people to notice when someone is struggling, but with no structure for what to do next. So they improvise. They end up absorbing risk they were never equipped to carry and have no clean way to escalate. The hazard stays in one person's head until it becomes a claim.

What sits upstream of the EAP?

This is the part nobody has built, because it falls between functions. It is not the EAP's job — that activates after distress. It is not resilience training, which puts the entire load on the individual to cope better with a hazard you have not controlled. And it is not a one-off awareness day. It is structured prevention: a wellbeing infrastructure that sits in the operating model, owned by operations, not bolted on by HR.

In practice that means treating psychosocial risk the way you treat any other risk. A real risk assessment that names the hazards specific to your sites and roles. A risk register that puts them on the same page as your physical risks, with owners and controls. A psychosocial safety framework that gives managers a defined pathway — what to notice, what to do, when and how to escalate — so the work no longer depends on whichever supervisor happens to be observant. Wellbeing champions with a structure behind them, not goodwill in front of a void.

That is the layer the Pathway builds. And here is the part that resolves the apparent contradiction in your data. Build it and EAP utilisation does not just rise — it gets healthier. People are routed to the right support earlier, when it is appropriate, instead of arriving at crisis. At the same time the absolute number of people who escalate to crisis falls, because the hazard is being managed before it compounds. You increase appropriate use of the EAP and reduce dependence on it. Both, at once. That is what an upstream control does that a downstream one structurally cannot.

This is the buying decision underneath the buying decision. The EAP renews every year and the claims keep coming, because you are paying to manage the consequence. The problem bills you every quarter. The infrastructure that closes the gap bills you once and changes what the next utilisation report — and the next claims line — actually says.