The week-two conversation that prevents the week-twenty claim

The post-incident review reads the same way nearly every time. The indicators are often dispersed across the crew. One colleague notices the person has become unusually quiet. Another observes a rise in absences. A third recalls uncharacteristic irritability and a tendency to snap at others over minor frustrations. Individually, each sign may seem insignificant; together, they form a pattern of emerging psychosocial risk. The pattern was sitting in plain sight at floor level for weeks. And then the line that should make every operator uncomfortable: "They seemed fine three days ago." Both things are true. The signals were visible, and nobody escalated them, because nobody had a way to.

This is the question worth sitting with: not "why didn't the EAP catch this", because an EAP is reactive support that activates after someone self-identifies and reaches out. The question is why a fortnight of observable change at crew level never travelled the three metres to a supervisor who could act on it. That gap — between what crews are noticing and what the organisation is able to recognise, interpret, and act upon — is where psychosocial risk often takes hold. Yet in many organisations, it remains a blind spot, with no formal mechanism to convert observation into early intervention.

Why does distress stay invisible to the people paid to manage it?

It does not stay invisible to everyone. The people most likely to notice a problem are rarely the ones responsible for responding to it. Those working alongside someone every day often spot the changes first: the normally talkative colleague who has gone quiet, the reliable worker who is suddenly making mistakes, the withdrawal, fatigue, irritability, and missed days. The signs are not hidden. They are simply observed by the workforce but rarely translated into action by the organisation.

Dunbar's number is useful here. A person can maintain around 150 stable relationships, but only a much smaller number of people are known well enough for any changes in behaviour to be noticed quickly. Supervisors and crew leaders already sit within that circle for their teams. They have the observational data. What they often lack is the permission, language, and escalation pathway to act on it. The result is predictable. A colleague becomes an accidental counsellor, carrying the situation informally with no training or support, or the changes are quietly normalised until they are no longer seen as warning signs at all. By the time the issue reaches formal attention, it has often become a crisis.

The change is measurable before it becomes a claim. Burnout is strongly predicted by perceived control, workplace change, and psychological experience at work (CMHAA, 2025). These risks emerge first in behaviour — withdrawal, fatigue, conflict, and performance changes — long before they appear in claims data. Among employees rating their own performance below their usual level — and that is 45% of them — 60% report symptoms of burnout (CMHAA, 2025). Presenteeism is not a productivity footnote. It is a leading indicator, visible to crews weeks before it becomes visible to organisational reporting systems.

What does "they seem okay" actually cost when it's wrong?

Consider it a risk register entry. Serious psychological injury claims result in a median 35.7 weeks of lost time, compared with 7.4 weeks for other serious claims (Safe Work Australia, cited in SafetySure, 2025). The cost appears late; the warning signs appear early. That is not a marginally worse outcome. It is roughly five times the disruption — the position held open, the backfill, the productivity drag, the premium exposure - and the WHS obligations that now attach to psychosocial hazards as a workplace risk requiring control under the hierarchy of controls, not merely a wellbeing initiative.

The compounding factor is work stress. Among employees carrying both a mental or emotional concern and work stress, 88% are in burnout, against 11% of the no-stress group (CMHAA, 2025). Work conditions are not a neutral backdrop to personal difficulties; they are an active amplifier that organisations can influence. This is not an HR awareness issue. It is an operating-model issue and an officer due-diligence issue. The reasonably practicable test asks what a reasonable officer would have done with what was knowable. In many cases, the risk was knowable. What was missing was a mechanism to turn observation into action.

What if your supervisors could spot trouble at week two instead of week twenty?

TThis is a design problem, not an awareness problem. You do not need every supervisor to become a counsellor — that model does not scale and it does not last. You need a small number of trained observers on each site and a clear, repeatable action pathway that sits between "they seem okay" and "call the EAP"..

The trained eyes are wellbeing champions: people already embedded in their crews, given a real method instead of a poster. The method has to be simple enough to use under operational pressure. We run it as SCOPE:

• Scan — notice the change against the person's own baseline, not a generic checklist of symptoms.
• Check in — a direct, low-stakes conversation that a trained person knows how to open and, crucially, when not to.
• Open space — give the conversation room rather than rushing to fix or refer.
• Prompt next step — connect the person to the right support, which is often, but not always, the EAP.
• Escalate — a defined path so the observation reaches someone who can act, with the floor-level signal preserved instead of lost.

This is the wellbeing infrastructure that sits before the EAP — prevention at the top of the cliff rather than intervention at the bottom. It does two things at once: reduces the number of issues that escalate into claims by catching them early, and increases appropriate EAP uptake by helping people access support before they reach crisis. Fewer escalations and better uptake, from the same intervention.

How is this different from what we already pay for?

Your EAP is necessary and you should keep it. But it is downstream by design — it activates after harm, on self-referral, once someone has already realised they are not coping. Resilience training does not close the gap. It asks the individual to absorb more load rather than equipping the organisation to detect when that load is building. Mental health awareness alone does not close the gap. It gives people a vocabulary for concern without providing a pathway for action, often creating accidental counsellors who feel responsible for what they have noticed but have nowhere appropriate to escalate it.

That is the Pathway: the infrastructure that ensures a week-two signal does not have to wait until week twenty to matter.