When is health monitoring required? Safe Work Australia says its own threshold is too vague to apply

Updated 17 August 2026, later the same day. The first version took Western Australia’s arrangement from Safe Work Australia’s description of it, and said in our view section that the report count WA receives would settle the question. We have since read WorkSafe WA’s own guidance for registered medical practitioners and added a section on what copying WA would actually mean, which is more than a reporting rule. We also went looking for the report count and could not find it published; the view section now says so. Nothing in the original was withdrawn.

Updated again 17 August 2026. We obtained and read the Work Health and Safety Commission’s 2024-25 annual report, the last public document we thought might carry the WA report count. It does not mention health monitoring, so the view section now says that too. The count does not appear to be published anywhere public.

The two things, and where the duties sit

They are routinely spoken of as a pair and they are not the same duty. Health monitoring is a registered medical practitioner monitoring a worker for changes in health status caused by exposure. Air monitoring measures the air, to establish whether airborne concentrations exceed the workplace exposure limit. One tells you a person has been affected; the other tells you the control is failing before anyone is.

In the model WHS Regulations, air monitoring sits at regulations 49 and 50 in Part 3.2, plus regulation 352; health monitoring sits at regulations 352 and 368 to 378 in Part 7.1, with Schedule 14 listing the chemicals that require it and the type required. Lead, crystalline silica and asbestos carry their own health monitoring rules in Parts 7.2, 8A.1 and 8.5. Above all of it sits section 19(3)(g) of the model Act, the duty to ensure so far as is reasonably practicable that worker health and workplace conditions are monitored to prevent illness or injury.

The threshold nobody can apply

Regulation 368 sets out when a PCBU must provide health monitoring, and one limb of it turns on whether there is a significant risk to a worker’s health from exposure during ongoing work. The paper’s account of why that does not work is blunt: the term ‘significant risk’ is not used elsewhere in model WHS legislation and is not defined in the Regulations. It adds that PCBUs may decide the threshold is not met and so delay monitoring or never do it, and that different PCBUs read it differently. The guidance was supposed to close the gap; the paper says stakeholders report it has not.

The options are the two you would expect, and the second is more radical than it looks. Option 1.1 keeps the words and defines them in the Regulations. Option 1.2 deletes significant and leaves plain risk, which would move the trigger from a judgement about severity to the ordinary risk language used everywhere else in the laws. Three of the other health monitoring issues are the same shape: valid techniques, ongoing work and biological exposure standard are all terms the Regulations use and do not define.

Every issue and option in the consultation paper, as the paper itself numbers them.

Health and air monitoring review, issues and options

No.	Issue	Options
1	Threshold: definition of “significant risk”	Define it in the Regulations, or replace it with “risk”
2	Threshold: risk of exposure versus risk to health	Amend so the duty applies to a risk to health because of exposure, or improve guidance
3	Threshold: “valid techniques”	Define the term, or improve guidance
4	Threshold: “ongoing work”	Define it and capture a broader range of workers, or improve guidance
5	Biological exposure standards	Define the term, or improve guidance
6	Schedule 14, the list of chemicals requiring health monitoring	New criteria, then either retain and update Schedule 14 or remove it for a separate list
7	Who sends the health monitoring report to the regulator	The doctor sends it; the doctor sends it and the threshold drops; or the threshold drops and the PCBU still sends it
8	Information given to the medical practitioner	Require air monitoring reports be provided to them, or improve guidance
9	Linking monitoring to risk management	Require controls be revised in response to health monitoring, or improve guidance
10	Clarifying the air monitoring requirements	Amend regulations 49 and 50, or clarify them in the model Code
11	Disconnect with other hazardous chemical requirements	Integrate or cross-reference Part 7.1, update the model Code, or improve other guidance

Issue 7 is the one to read: Western Australia already does it differently

Regulation 376 requires the PCBU to send a worker’s health monitoring report to the WHS regulator, and only where the report indicates the worker may have contracted a disease, injury or illness, or recommends remedial action. Regulators, the paper says, receive relatively few of them, which is limiting visibility of what health monitoring is occurring, how often health monitoring is undertaken and where workers may be exposed.

Western Australia’s version of regulation 376 differs in two ways, and both matter. The registered medical practitioner, not the PCBU, submits the report. And the trigger is lower: a report goes to the regulator if results are consistent with exposure to a hazardous chemical, rather than only when they indicate disease or injury.

The paper reports the result plainly. The WA approach has reportedly produced more reports, improving visibility of workplace exposures and supporting earlier intervention and more proactive regulatory oversight, and has encouraged greater participation in health monitoring by PCBUs, workers and practitioners alike. Options 7.1, 7.2 and 7.3 are, in effect, how much of the WA model to adopt: the doctor reporting, the lower threshold, or both. The paper is even-handed about the cost, noting that it may increase the administrative responsibilities of medical practitioners and require more resourcing for regulators.

What copying Western Australia would actually mean

The consultation paper describes WA’s difference as a reporting rule. WorkSafe WA’s own guidance for registered medical practitioners describes something larger: the rule sits inside an apparatus the model laws do not have, and a submission arguing for or against option 7 is really arguing about that apparatus.

Start with the trigger, which is broader than the paper’s summary conveys. WorkSafe WA tells practitioners regulation 376 requires them to give the regulator a copy of the report if the results are consistent with exposure (this includes where the work tasks involve exposure). The parenthesis is doing real work: it reaches reports where the tasks involve exposure, not only where a test result says so.

Then the parts that are not in the model laws at all:

• It is an offence. WorkSafe WA states that Contravention of regulation 376 may result in a penalty of $7,000 for an individual or $35,000 for body corporate. The duty to report is enforceable against the doctor.
• The form is prescribed. The RMP is required to use the health monitoring report forms, so what arrives at the regulator is consistent enough to count.
• There is a phone number for the bad case. On a medical removal the practitioner notifies WorkSafe promptly by phone and examines the worker before recommending a return to that chemical. That is a same-day channel, not a form.
• There are clinicians on the other end. The RMP may request to speak to the WorkSafe Occupational Health Nurse or the Occupational Physician to clarify an issue. The regulator staffs the relationship.
• WA writes its own practitioner guides and points to Safe Work Australia’s only in their absence.

WorkSafe WA also puts a number on the list the consultation is reconsidering: Schedule 14 Table 14.1 covers 15 main hazardous chemicals for which health monitoring is required on a risk basis, with Table 14.2 for lead and Part 8.5 Division 1 for asbestos.

None of that settles whether the national rule should change. It does mean the question in front of stakeholders is narrower than it looks: option 7 moves the reporting duty, and moving a duty onto practitioners who have no prescribed form, no clinician to call and no penalty attached is not the same intervention that produced WA’s results. That is a point a submission can make with WA’s own guidance as the evidence.

Air monitoring: two regulations pulling against each other

Regulation 49 requires a PCBU to ensure that no person at the workplace is exposed to a substance or mixture in an airborne concentration that exceeds the exposure standard for it. Regulation 50 requires air monitoring where the PCBU is not certain on reasonable grounds whether the standard is exceeded, or where monitoring is needed to determine a risk to health.

Read together, they raise a question the paper says stakeholders keep asking: whether there is an obligation to monitor every worker at all times to guarantee no exceedance. And regulation 50’s escape hatch, certainty on reasonable grounds, is undefined in practice. Can you rely on old monitoring data? Third-party data? Statistical sampling rather than every worker? The paper says these questions bite hardest for small and medium businesses and workplaces without specialist expertise, and notes that regulation 50 does not expressly require air monitoring to be done by a competent person.

How much exposure there actually is

The paper carries figures from Safe Work Australia’s forthcoming 2026 Australian Worker Exposure Survey, which estimates probable exposure from workers’ self-reported tasks and controls. Of surveyed workers, 68 per cent had probable exposure to at least one agent assessed. Within that group, 31.2 per cent were exposed to solvents, 17.5 per cent to dusts, 14.3 per cent to chemicals and 8 per cent to metals. The most common individual exposures were styrene at 9.1 per cent of workers and formaldehyde at 6.5 per cent. High-level formaldehyde exposure concentrates in construction (9.3 per cent), manufacturing (5.3 per cent) and wholesale trade (4.2 per cent).

Against that, the measured harm is small and almost certainly understated. The AIHW’s Australian Burden of Disease Study 2024 put occupational exposures and hazards at 1.6 per cent of the total burden of disease. The paper points out why compensation data cannot fill the gap: the ABS estimated that in 2021-22 only 31 per cent of the 497,300 people who experienced a work-related injury or illness received workers’ compensation. Internationally the ILO reports that reliable exposure data is unavailable for the majority of hazardous chemicals, so the burden is likely substantially underestimated.

Our view, labelled as such

Two things are worth saying, and the first is a credit. A regulator publishing that its own drafting is too unclear to apply, and printing the count of terms it uses without defining them, is doing something useful. Four of the eleven issues are literally undefined words in a binding regulation. That is not a discovery a duty holder could have made and acted on; it is one only the rule-maker can fix.

The second is where we would push. Issue 7 is the only proposal in the paper that changes what the regulator can see, and it has a working precedent in a jurisdiction that has run it for years. The paper describes WA’s results as reported rather than measured, and that is honest, but it also means the strongest evidence in the document is the softest sourced. If the case for the WA model rests on it producing more reports, the number of reports WA receives against the other jurisdictions would settle the question rather than leave it to submissions. We went looking for that number and could not find it published. WorkSafe WA’s data and statistics pages carry no count of health monitoring reports received, and the Work Health and Safety Commission’s 2024-25 annual report, which we read in full, does not mention health monitoring at all. Somebody holds the number; publishing it would be worth more to this consultation than most submissions.

There is also a sequencing problem worth naming. Better definitions raise how often monitoring is required; the reporting change raises how much of it the regulator sees. Adopt the first without the second and the visibility gap stays exactly where it is, because nobody will be able to tell whether the definitions worked. The paper does not link the two, and a submission could.

If you want to be in it

Submissions close at 11:59pm AEST on Monday 21 September 2026, through the Consultation Hub or by email to Safe Work Australia’s occupational hygiene address. The paper carries 26 discussion questions: 1 to 20 on health monitoring, 21 to 24 on air monitoring, and 25 to 26 for anything else. You may answer some or all of them, and Safe Work Australia asks for evidence or data with your view. You choose whether your submission is published under your name, your organisation’s, a pseudonym, anonymously, or not at all.

Anything agreed goes to Safe Work Australia Members as recommendations, then to Commonwealth, state and territory WHS ministers, who are the decision makers on the model laws. Nothing in this paper changes a duty today. We will report what the Members recommend.