Every Australian jurisdiction now requires employers to identify, assess, and control psychosocial hazards. With the commencement of Victoria’s Occupational Health and Safety (Psychological Health) Regulations on 1 December 2025, the national transition is substantially complete for psychosocial regulation in Australia. Psychological health now sits alongside physical safety as an explicit, enforceable regulatory obligation rather than an implied component of a general duty. But it’s a challenging space and organisations are trying to come to terms with the arguments for and against.
The shift to psychosocial regulatory reform did not happen quietly in Australia. Supporters typically describe it as the most significant work health and safety reform in a generation. Critics describe a compliance burden built on a range of concepts, enforced inconsistently, and producing paperwork rather than prevention.
This article examines the evidence based and it asks three fundamental questions in sequence. First, is there a problem of sufficient scale and severity to justify regulatory intervention? Second, is the problem genuinely work-caused, such that workplace regulation is the right instrument? Third, does regulation demonstrably change the outcomes, or would guidance and voluntary action achieve the same result?
The scale of psychosocial harm | what the claims data shows
The starting point for any regulatory case is defining and examining the harm data, and the Australian data is unambiguous.
Safe Work Australia’s Key Work Health and Safety Statistics for 2025 record 17,600 serious workers’ compensation claims for mental health conditions in 2023-24, an increase of 14.7 per cent on the previous year’s 15,300 claims. Mental health conditions now account for 12 per cent of all serious claims, the highest proportion on record and a 161 per cent increase compared with ten years earlier, the largest growth of any injury category, with median time lost almost five times longer than for other injuries and diseases (Safe Work Australia 2025).
The severity profile is more striking than the volume of claims. The median compensation paid for a serious mental health condition claim in 2022-23 was $67,400, more than four times the $16,300 median across all serious claims (Safe Work Australia 2025). Safe Work Australia’s dedicated analysis of psychological health in the workplace adds the human dimension which shows that workers with mental health condition claims experience poorer return to work outcomes and higher rates of stigma from colleagues and employers than workers with physical injuries (Safe Work Australia 2024).
The hazard profile behind those psychosocial claims matters for the regulatory question. The most common causes of serious mental stress claims in 2023-24 were harassment and workplace bullying (33.2 per cent), work pressure (24.2 per cent), and exposure to violence and harassment (15.7 per cent) (Safe Work Australia 2025). These are not random misfortunes. They are conditions arising from work design, management practice, and organisational climate, which is to say they are conditions a person conducting a business or undertaking can influence.
The most widely quoted international figures require more careful handling, because they were produced in an advocacy context. The World Health Organization and the International Labour Organization estimate that 12 billion working days are lost globally each year to depression and anxiety, at a cost approaching US$1 trillion in lost productivity (WHO and ILO 2022). These are modelled estimates, derived from return-on-investment analysis of scaled-up depression and anxiety treatment (Chisholm et al. 2016), and they count all depression and anxiety among workers regardless of cause.
They typically describe the scale of mental ill-health in working populations, not the harm attributable to work, and this article does not rely on them for the regulatory case. The more evidentially useful figure in the same policy brief is its attribution estimate where approximately 11 per cent of depression worldwide is attributable to occupational risks. That figure works in both directions. It confirms a material work-caused burden while conceding that most depression arises outside the workplace, which is a constraint any honest regulatory argument should accept.
Direct national surveillance is likely a more instructive measure. In Great Britain, where there are no psychosocial-specific regulations and the regulator relies on guidance under the general duty, the Health and Safety Executive’s 2024-25 statistics recorded 964,000 workers suffering work-related stress, depression or anxiety and 22.1 million working days lost, accounting for 62 per cent of all working days lost to work-related ill health and representing a rate that has more than doubled since records began in 2001-02 (HSE 2025).
For the Australian economy as a whole, the Productivity Commission’s Mental Health Inquiry estimated the total cost of mental ill-health at $200 to $220 billion per year and identified mentally healthy workplaces as one of five priority reform areas (Productivity Commission 2020). At the employer level, analysis commissioned by Safe Work Australia estimated the cost of depression alone to Australian employers at approximately $6.3 billion per annum through absenteeism and presenteeism (Becher and Dollard 2016). The Queensland Government’s explanatory material for its 2022 psychosocial regulation put the annual cost of work-related psychological injury and illness at $4.7 billion in absenteeism, $6.1 billion in presenteeism, and $146 million in compensation claims (Queensland Government 2022).
A fair reading of this data supports two conclusions. The harm is large, growing, and expensive. It is also concentrated in causes that sit within organisational control.
Does the causal evidence show the harm genuinely work-caused?
Claims data alone cannot justify regulations in practice. Compensation statistics measure what is claimed and accepted, not what work causes, and they are sensitive to awareness, scheme design, and reporting behaviour. The regulatory case therefore depends on the epidemiological evidence, and that evidence base is now substantial and growing rapidly.
The most studied exposure is job strain, the combination of high demands and low control first formalised in Karasek’s demand-control model. A systematic review and meta-analysis by the IPD-Work Consortium, drawing on six published cohort studies covering 27,461 individuals and 914 incident cases, found job strain associated with an increased risk of clinically diagnosed depression, extending earlier findings that had relied on symptom measures (Madsen et al. 2017). A meta-review of 72 literature reviews with meta-analysis concluded that associations between psychosocial work factors and both mental disorders and cardiovascular disease were generally significant, with the strongest and most consistent findings for job strain and long working hours in relation to coronary heart disease, stroke, and depression (Niedhammer, Bertrais and Witt 2021).
The population-level burden estimates are equally relevant to the regulatory question, because regulation is a population-level instrument. European attributable fraction analysis across 35 countries found that 17 per cent of depression was attributable to job strain, with further significant fractions for job insecurity (9 per cent), bullying (7 per cent), and effort-reward imbalance (6 per cent); the overall fraction of depression attributable to psychosocial work exposures ranged between 17 and 35 per cent (Niedhammer et al. 2021). In Australia, LaMontagne and colleagues applied population attributable risk methods to Victorian exposure data using a meta-analytic odds ratio of 1.82 for job strain and depression, concluding that job strain-attributable depression represents a substantial, preventable, and inequitably distributed public health problem, and that the volume of compensated mental stress claims substantially understates the true burden (LaMontagne et al. 2008).
Two features of this literature deserve acknowledgement in that most studies are observational, and residual confounding and reporting bias cannot be fully excluded. The meta-review authors themselves note that causal inference remains difficult (Niedhammer, Bertrais and Witt 2021). The effect sizes are also moderate rather than dramatic. Subsequently the counterpoint, made explicitly in the systematic review literature, is that when an exposure is as prevalent as job strain, even a moderately elevated relative risk produces a large population attributable risk. A work environment factor with 25 per cent prevalence and a relative risk of 1.8 yields a population attributable risk of around 11 per cent (Theorell et al. 2015).
Common exposures with moderate risks are precisely the territory in which public health regulation has historically operated, from noise to manual handling.
The conclusion this evidence supports is likely measured. Psychosocial hazards at work are clearly established risk factors for depression, anxiety, and cardiovascular disease. The exposures are potentially modifiable through work design and the burden is large because the exposures are typically common in workplaces This is ultimately the classic profile of an occupational hazard that requires regulation.
How Australia arrived at regulation of psychosocial risks
The Australian regulatory response was not a reaction to a single event which is not uncommon from a regulatory perspective. The ultimate outcome was a decade of converging findings on the growing prevalence of psychosocial risk in Australian society.
The Safework Australia model WHS Act has always defined health to include psychological health but what the framework lacked, until 2022, was any regulation or code explaining how the duty applied to psychosocial hazards. Marie Boland’s 2018 independent review of the model WHS laws identified this gap directly and in her second recommendation, of 34 recommendations in total, was that the model WHS Regulations be amended to deal with how to identify psychosocial risks and the appropriate control measures to manage them. Subsequently she reported that the need for more sound architecture around psychosocial risk was one of the top three issues raised by business owners, workers, and regulators alike throughout her national consultations (Boland 2018; Safety Solutions 2019).
The recommendation drew support from the union movement, WHS academics, and professional bodies, and notably, the demand for clearer architecture came from duty holders as much as from worker representatives.
The Respect@Work national inquiry into workplace sexual harassment (2020) and the Productivity Commission’s Mental Health Inquiry (2020) reinforced the same conclusion from different directions:. Existing settings were not preventing psychological harm at work, and clearer positive obligations were needed to bring about control of this emerging issue. WHS ministers agreed to implement the Boland recommendation in 2021, the model WHS Regulations were amended in 2022, inserting regulations 55A to 55D, which define psychosocial hazards and risks and require control through work design, systems of work, and the work environment so far as is reasonably practicable. The model Code of Practice, Managing Psychosocial Hazards at Work, followed in July 2022.
Queensland commenced its mirror regulation on 1 April 2023 and went further than the model by applying the hierarchy of controls to psychosocial risks. New South Wales had moved earliest, commencing its psychosocial amendment regulation on 1 October 2022, and its Work Health and Safety Regulation 2025 now routes psychosocial risk management through the general risk management path, picking up the hierarchy of control measures. Victoria, outside the harmonised scheme, completed the national picture with its standalone psychological health regulations commencing 1 December 2025.
The sequence matters for the evidence questions that the author has asked. Regulation did not precede the evidence, it simply followed approximately fifteen years of Australian surveillance research, two national inquiries, and an independent statutory review, each of which independently concluded that guidance alone had not closed the gap to mitigate the affects of psychological injury in the workplace.
The complexity of applying the psychosocial codes in practice
The codes of practice fundamentally solved the architecture problem Boland identified at the level of duty. What is less often acknowledged is that they transferred much of the complexity to the point of application, and that complexity operates on several levels at once.
The same code of practice carries three different legal effects
A national employer applying “the psychosocial code” is in fact applying a mandatory benchmark in Queensland, an evidentiary benchmark in the harmonised states, and an optional safe harbour in Victoria. Section 26A of the Work Health and Safety Act 2011 (Qld) requires a PCBU to “comply with the code” or to manage the relevant hazards and risks “in a way that is different to the code but provides a standard of health and safety that is equivalent to or higher than the standard required under the code”.
Departing from the Queensland code therefore demands a documented equivalence case. In New South Wales, by contrast, s 275 of the Work Health and Safety Act 2011 (NSW) makes an approved code admissible as evidence, and the court “may… rely on the code in determining what is reasonably practicable”, which is persuasive force rather than obligation, although from 1 July 2026 New South Wales moves to the Queensland model, requiring duty holders to follow approved codes unless an equivalent or higher standard is demonstrated. Victoria inverts the logic again: under s 150 of the Occupational Health and Safety Act 2004 (Vic), “a failure to comply with a compliance code does not give rise to any civil or criminal liability”, while s 152 deems a person who complies with a compliance code to have complied with the corresponding duty to the extent the code covers it. One hazard profile, three compliance postures. Multi-state organisations cannot run a single compliance model off the code text alone.
The hierarchy of controls exists in four versions
The model regulations deliberately omitted the hierarchy for management of psychosocial risk, requiring instead that control measures have regard to “all relevant matters” listed in r 55D. Queensland applied the standard hierarchy to psychosocial hazards. The NSW Work Health and Safety Regulation 2025 reaches it by a different route: r 55C requires psychosocial risks to be managed in accordance with the general risk management part, which contains the r 36 hierarchy, while r 55D imposes the eliminate-then-minimise duty and the relevant-matters list.
Victoria built its own: r 15 of the Psychological Health Regulations requires elimination so far as is reasonably practicable, then risk reduction by altering the management of work, plant, systems of work, work design, or the workplace environment, and permits information, instruction or training as the exclusive control only “if none of the control measures set out in subregulation (2)(a) are reasonably practicable”, with the further restriction that in any combination such measures “must not be the predominant control measure used”.
Victoria has, in effect, legislated against the control most organisations reach for first. The hierarchy also translates awkwardly in concept | eliminating a chemical is a coherent instruction, whereas eliminating job demands, organisational change, or exposure to distressed people frequently is not, and the codes offer limited help converting engineering-control logic into work design decisions.
The codes regulate the work, not the worker
Perhaps the deepest conceptual difficulty is that psychological harm emerges from the interaction between working conditions and the individual, yet the general provisions are framed entirely around the work. The matters a PCBU must consider under r 55D are the duration, frequency and severity of exposure, how hazards interact or combine, the design of work, systems of work, the workplace environment, plant, workplace interactions, and training and supervision.
Individual characteristics and individual responses do not appear in any Australian code of practice. The omission is clearly a deliberate design choice with defensible reasons in that the duty attaches to what the PCBU controls, and a framework built around individual susceptibility would invite screening of “vulnerable” workers, with the discrimination and privacy problems that follow.
The framework is not entirely blind to the issue but recognises that multi causal factors contribute. Queensland’s sexual and gender-based harassment subdivision, inserted in 2024, requires control measures to have regard to “characteristics of the workers”, including age, gender, sex, sexual orientation, and disability (r 55F, WHS Regulation 2011 (Qld)), and known or foreseeable vulnerability remains relevant to what is reasonably practicable under the general duty.
But for the general run of psychosocial hazards, WHS practitioners are asked to manage a person-environment interaction using instruments that describe only the environment, while the compensation system on the other side of the ledger responds to individual outcomes however atypical the response. The science sits uncomfortably between the two for example the same workload that harms one worker engages another, and exposure-based assessment tools cannot fully resolve that variance.
Some psychosocial risk exposure is inherent to the occupation
The difficulty sharpens in work where the hazard is the job in itself. For example a mental health nurse cannot be insulated from acutely unwell, distressed, or aggressive patients without ceasing to be a mental health nurse, and the same holds for paramedicine, child protection, corrections, and emergency services.
Elimination-first logic typically strains badly here, and practitioners in these sectors are entitled to ask what the codes expect of them.
Two responses are available, and both matter significantly to the issue at hand. The first is that inherent danger has never excused the control duty for physical hazards such as mining and firefighting did not escape regulation because the danger was intrinsic, and the reasonably practicable calculus absorbed the inherent nature of the work by shifting the emphasis from elimination to minimisation through design such as staffing models, environmental and duress design, rostering and recovery, clinical supervision, and structured post-incident support are the psychosocial equivalents.
The second response is empirical and psychosocial safety climate research demonstrates that organisations with comparable inherent exposure produce materially different harm outcomes depending on the organisational climate for psychological health, with high-climate workplaces buffering the relationship between demands and harm (Dollard and Bakker 2010). The concentration of psychological claims in health care and social assistance is therefore not simply the cost of intrinsically hard work. Part of it reflects a long history of treating harm as part of the job (normalisation of risk potentially), and the codes, whatever their limitations, exist to dismantle exactly that normalisation.
These application difficulties are real, and organisations and regulators are potentially still working through them and what they mean. They are, however, the same problems encountered in the first decade of every regulated hazard class such as divergent jurisdictional drafting, immature assessment methods, and contested boundaries. They define the frontier where the psychosocial codes will need to mature, particularly on cumulative exposure, individual variability, and inherent-exposure occupations. They do not unsettle the underlying case for the duty.
Does regulation actually work? The evidence for effect
This is the hardest question, and the one where advocates on both sides most often substitute assertion for data.
Regulation is the dominant driver of employer action
The European Survey of Enterprises on New and Emerging Risks (ESENER), the largest recurring dataset on how establishments manage workplace risk, has found across successive waves that fulfilling a legal obligation is the leading reported driver of health and safety action. In ESENER-3, 89.2 per cent of responding employers nominated fulfilment of a legal obligation as the main reason for addressing health and safety, and the 2024 wave recorded 87 per cent (Eurocadres 2020; ETUI 2025). EU-OSHA’s dedicated analysis of psychosocial risk management reached the same conclusion: legal obligation is a powerful driver of psychosocial risk management specifically, and non-binding policies work best where they make the link between the legal case and the business case explicit (EU-OSHA 2012). If the objective is to move the majority of employers rather than the motivated minority, the survey evidence indicates that law is the lever that moves them.
Targeted psychosocial law is associated with measurable improvement
Direct evaluation evidence is recent, because the laws are recent, but it is beginning to trickle down. A multi-study policy evaluation published in Safety Science by Potter, Dollard, and colleagues combined expert reports from 19 countries confirming Australia’s increased legislative focus, a quasi-experimental comparison of national datasets showing significant increases in psychosocial safety climate in covered jurisdictions, and a three-wave longitudinal sector study showing sustained improvement in psychosocial safety climate and reductions in psychological distress following the introduction of the new WHS regulations.
The authors concluded that targeted psychosocial law had a positive impact on organisational safety systems and worker health, and that legislation can operate as a wide-reaching lever for improving working conditions over the longer term (Potter et al. 2025). One evaluation clearly does not settle the question but it is, however, precisely the kind of evidence the sceptics said did not exist, and its direction is consistent with the ESENER findings.
The interventions regulation requires are the interventions that work
The regulations do not mandate wellness programs or resilience training.
They mandate risk management directed at work design, systems of work, and the work environment, which is to say organisational-level intervention.
The intervention evaluation literature supports that emphasis and LaMontagne and colleagues’ systematic review of the job stress intervention literature found that individually focused approaches improve individual outcomes but tend not to change organisational ones, while approaches incorporating organisational change deliver benefits at both levels (LaMontagne et al. 2007). Another 2023 overview of 52 systematic reviews covering 957 primary studies found strong quality evidence for organisational interventions targeting working time arrangements and burnout, and moderate quality evidence for changes to work tasks, work organisation, and the psychosocial work environment (Aust et al. 2023).
The economic appraisals point the same way, if modestly by example the Deloitte Access Economics regulatory impact statement prepared for the Victorian regulations estimated that benefits from improved productivity and reduced claims would exceed compliance costs, at a reported ratio of 1.06 to 1 (Deloitte Access Economics 2022).
In Australian organisational research, the psychosocial safety climate construct developed by Dollard and Bakker provides the connecting theory in that the organisational climate for worker psychological health functions as a leading indicator of working conditions, worker health, and engagement, and Australian Workplace Barometer research has linked low psychosocial safety climate to billions of dollars in annual productivity loss (Dollard and Bakker 2010; Becher and Dollard 2016).
Regulation that obliges senior management to prioritise psychological health is, in effect, regulation that targets the leading indicator.
The case against psychosocial regulation
A balanced evidence review must give the counter-arguments their full weight, and several survive scrutiny better than advocates of regulation tend to admit.
The subjectivity objection
Psychosocial hazards include concepts such as poor organisational justice, low role clarity, and inadequate reward and recognition. Critics argue that what constitutes justice or adequate support depends heavily on individual perception of these issues, and that recognising these as regulated hazards risks converting ordinary workplace unhappiness into compliance exposure, with every management decision open to challenge (Squire Patton Boggs 2025). The concern is not unwarranted in that physical hazards are measurable eg a noise dose or an airborne concentration can be objectively established. Psychosocial exposure assessment relies substantially on self-report, and validated instruments mitigate rather than eliminate that limitation. The strongest response is that the duty attaches to the system of work, not to individual feelings in that a PCBU is judged on whether foreseeable hazards were identified and reasonably practicable controls implemented, not on whether every worker felt content.
The model Code’s recognition of reasonable management action carried out in a reasonable way preserves the employer’s capacity to manage performance and conduct. The objection nonetheless identifies a real boundary-drawing problem that regulators and courts are still working through as this article is presented.
The justiciability objection
Enforcement experience to date gives this objection some genuine force. Recently SafeWork NSW withdrew its prosecution of Western Sydney Local Health District after three weeks of hearing and twenty witnesses, in a matter alleging failures in the management of complaint and grievance processes affecting two nurses.
The court observed during the proceedings that complaint handling is inherently stressful and that causing stress through such processes is not of itself a breach of the WHS Act (Ashurst 2025; SafeWork NSW v Western Sydney Local Health District (No 3) [2025] NSWDC 48).
Commentators reasonably concluded that proving psychosocial breaches to the criminal standard is inherently difficult. Against that sit the successful outcomes viewed in WorkSafe Victoria’s conviction of Court Services Victoria over psychosocial harm to court staff, the Commonwealth Department of Defence’s conviction and $188,000 fine following the suicide of a RAAF technician exposed to high-pressure performance management at Williamtown. A steady escalation of enforcement activity including SafeWork NSW’s prohibition notice halting a University of Technology Sydney restructure on psychological risk grounds in September 2025, hundreds of psychosocial non-compliance notices through 2025 compliance blitzes, and the recruitment of 20 dedicated psychosocial inspectors announced in March 2026 (Law Society Journal 2026; ReFresh 2026).
The honest summary is that prosecution is hard, notices and improvement-focused enforcement are operating at scale, and the deterrent architecture is still maturing. Difficulty of prosecution is an argument for better-drafted duties and better-resourced regulators; it is not, on its own, an argument that the hazard should be unregulated.
The compliance theatre objection
Perhaps the most substantively important criticism is that early compliance activity has produced surveys and risk registers rather than changed work. Practitioner analysis has documented risk registers swollen to hundreds of items, many of which record control failures rather than hazards, and surveys interpreted against arbitrary thresholds with no follow-through (Wellbeing Australia 2026).
This criticism should be accepted, and it cuts in an unexpected direction. The same pattern characterised the early years of physical risk regulation, when safety management systems were measured by documentation volume rather than risk reduction. The intervention literature is explicit that organisational-level change, properly implemented, works effectively, and that assessment without intervention simply does not. The policy implication is that regulators and advisers must hold organisations to the control step of the risk management cycle, not that the cycle should never have been mandated. Paper compliance is a failure of implementation, not a refutation of the underlying duty.
The claims interpretation objection
Rising claims do not straightforwardly materially demonstrate rising harm in workplaces. Legal practitioners currently observe that the growth reflects a combination of factors such as greater legal and policy focus, greater awareness of rights, the changing complexity of work, and broader socioeconomic stress spilling into workplaces (Law Society Journal 2026).
This is likely correct, and it is why the regulatory case in this article does not rest on claims data alone. The epidemiological burden estimates, which are independent of compensation behaviour, indicate that claims understate rather than overstate the true volume of work-attributable psychological injury (LaMontagne et al. 2008). Under-claiming, stigma, and poor return to work outcomes for psychological injury are documented in Safe Work Australia’s own analysis (Safe Work Australia 2024). If anything, the awareness effect means the schemes are belatedly capturing harm that was always present.
The burden and proportionality objection from employers
Compliance costs are real, and typically fall disproportionately on small and medium enterprises, and were a recurring theme in stakeholder research before the psychosocial reforms. Potter and colleagues’ qualitative study of 25 informed participants across policy, regulation, and industry found genuinely divergent views on how psychosocial risks should be addressed in legislation, how compliance should be established, and what role regulators should play (Potter et al. 2019). The available economic appraisal suggests net benefit, though narrowly (Deloitte Access Economics 2022), and the cost asymmetry in the claims data, with median psychological claims costing more than four times the all-claims median, indicates that the cost of inaction is also concentrated and growing.
Proportionality is best served by scalable guidance and regulator support for smaller businesses rather than by exempting the hazard, particularly given that psychosocial hazards are not confined to large organisations.
The individual-contribution objection
The most structurally interesting objection evidenced by the authors own experiences runs as follows – psychological harm is substantially driven by individual characteristics and non-work circumstances, yet employers can neither select for psychological suitability nor quarantine themselves from claims where work was only one contributor among several.
Each limb of that objection is likely factually true, and an honest review should state this. The epidemiology itself concedes the causal arithmetic if psychosocial work exposures account for 17 to 35 per cent of depression, the majority share arises from individual and non-work factors, and prospective research consistently identifies prior psychopathology, childhood adversity, and traits such as neuroticism as predisposing factors.
On the legal side, discrimination law confines pre-employment health assessment to capacity to perform the inherent requirements of the role, and the Queensland compensation scheme requires only that employment be “a significant contributing factor” to a psychiatric injury (s 32, Workers’ Compensation and Rehabilitation Act 2003 (Qld)), the stricter “major significant contributing factor” test having been repealed in 2019.
Employers, in short, likely absorb a multifactorial risk they can partially cause and cannot select against the risk. What defeats the objection is not its premises but its implied remedy. The screening alternative has been tested, and it fails. A systematic review of pre-employment and pre-duty screening among emergency workers found no proven efficacy despite its continuing widespread use (Marshall et al. 2017). The most direct Australian evidence, a nested case-control study matching 150 police officers with accepted psychological injury claims against 150 healthy colleagues, found no association between validated pre-employment personality and psychopathology measures and subsequent psychological injury (Marshall et al. 2020).
The military experience reaches the same conclusion across a century of historical analysis. Mass screening found false positives greatly exceeding true positives (Jones, Hyams and Wessely 2003), and follow-up of UK forces screened before the Iraq deployment found screening would not have prevented the subsequent psychological morbidity (Rona et al. 2006). The base-rate arithmetic explains why in that weak predictors of uncommon outcomes exclude mostly people who would have remained well while clearing many who later become unwell. The compensation scheme, for its part, already contains its own concessions to multifactorial causation, compensating aggravations only to the extent of the aggravation and excluding disorders arising from reasonable management action taken reasonably.
The objection therefore resolves into an explanation rather than a refutation simply because neither worker selection nor non-work causation is within the employer’s effective control, work design is the only lever the duty holder actually holds, and the psychosocial safety climate evidence indicates that pulling it changes outcomes even among vulnerable workers (Dollard and Bakker 2010). The integrated intervention model in the academic literature completes the picture by assigning workplaces a support role for mental illness regardless of cause, without making them causally responsible for it (LaMontagne et al. 2014). The regulation targets the work because the work is what can be governed.
The comparative objection
Finally, sceptics sometimes point to jurisdictions with long-standing general duties and ask why more law would help the case. Great Britain is the instructive case for this argument It has operated a guidance-based approach to work-related stress, through the HSE Management Standards, under a general duty framework for two decades, without psychosocial-specific regulations.
Its self-reported work-related stress, depression and anxiety has reached record levels, with rates more than double those of 2001-02 and accelerating since the pandemic (HSE 2025; British Safety Council 2025). A single comparison cannot isolate the effect of regulatory design from labour market and cultural differences. It does, at minimum, demonstrate that guidance within a general duty has not arrested the trend in a comparable economy, which is consistent with the Boland Review’s central finding that duty holders needed enforceable architecture, and with the ESENER finding that legal obligation, rather than guidance, is what drives employer action.
Weighing the evidence for regulatory intervention
The case for regulation rests on convergent evidence from independent sources like compensation data showing large, growing, and disproportionately severe harm concentrated in work-design causes; epidemiology establishing psychosocial exposures as modifiable risk factors with substantial attributable burden; survey evidence across tens of thousands of enterprises showing legal obligation to be the dominant driver of employer action on these risks; intervention research showing that the organisational-level controls the regulations require are the controls that work; early quasi-experimental and longitudinal evidence that targeted Australian psychosocial law has improved psychosocial safety climate and reduced distress; and a comparative example suggesting that guidance-only approaches have not contained the problem elsewhere.
The case against rests largely on implementation critique with issues like the subjectivity at the boundaries of the duty, the difficulty of criminal prosecution, the risk of paper compliance, interpretive caution about claims trends, compliance cost, the individual-contribution objection, and the genuine application complexities examined above, including jurisdictional inconsistency in the legal status of the codes, their silence on individual variability, and the strain of elimination-first logic in occupations where exposure is inherent to the work. Each of these criticisms identifies something real.
None of them, on examination, argues that the hazard is not work-caused, that the harm is not substantial, or that voluntary approaches would do better, and the strongest of them, the individual-contribution objection, resolves on the screening evidence into an explanation of why the duty targets the work. They argue that regulation must be drafted carefully, enforced intelligently, and implemented as genuine risk management rather than documentation. Those are conditions on success, not arguments against the enterprise.
The conclusion the evidence supports is therefore likely a qualified affirmative in point. Psychosocial safety does require regulation, on the same logic that physical safety required it and there is no doubt that the hazards are real, prevalent, organisationally controllable, and inadequately addressed by voluntary action. Regulation is therefore necessary and the evidence is equally clear that it is not sufficiently capable of working effectively at this point in time. The Australian framework will deliver its promised benefits only if organisations move past the survey-and-register stage to redesign the work itself, and only if regulators sustain the enforcement capability that gives the duty practical meaning.
What this means for organisations
For boards and officers, the due diligence obligation now unambiguously extends to psychological health, and the enforcement trajectory, from prohibition notices halting restructures to convictions over performance management processes, indicates that regulators will look at how work is organised and how management processes are conducted, not just at whether policies exist.
For WHS practitioners, the evidence reviewed here points to a clear practice standard. Identification should draw on validated tools and worker consultation rather than generic hazard lists.
Assessment should distinguish hazards from control gaps.
Controls should target work design, workload, role clarity, support structures, and the conduct of management processes, because that is where both the regulations and the intervention evidence place the leverage. Review should measure whether exposure and harm have actually reduced effects. Organisations that treat the psychosocial regulations as a documentation exercise will accumulate registers vs organisations that treat them as a work design discipline will likely reduce risk, and the economic evidence (at this point) indicates they will recover the cost of doing so.
The regulation of psychosocial safety is sometimes framed as a question of belief, as though it turned on whether one takes mental health seriously.
It is better framed as a question of evidence, and the evidence, weighed fairly and with its limitations acknowledged, comes down on the side of the reforms now in force across every Australian jurisdiction.
Related reading from Safetysure
For the practical compliance counterpart to this evidence review, see our psychosocial risk management guide. The claims data examined above is analysed in depth in Mental health workplace claims Australia: 2025 statistics and analysis. On governance, our analysis of officer due diligence examines what the courts now expect of boards, and our evidence-based framework for regulatory design considers what makes WHS regulators effective. Offence categories and penalties are summarised in WHS offence categories.
References
Ashurst (2025) SafeWork NSW withdraws landmark psychosocial risk prosecution, Ashurst Insights, March 2025.
Aust B, Møller JL, Nordentoft M et al. (2023) ‘How effective are organizational-level interventions in improving the psychosocial work environment, health, and retention of workers? A systematic overview of systematic reviews’, Scandinavian Journal of Work, Environment and Health, 49(5):315-329.
Becher H and Dollard MF (2016) Psychosocial safety climate and better productivity in Australian workplaces: costs, productivity, presenteeism, absenteeism, Safe Work Australia, Canberra.
Boland M (2018) Review of the model Work Health and Safety laws: final report, Safe Work Australia, Canberra.
British Safety Council (2025) Work-related stress, anxiety and depression reach record high, HSE stats show, 21 November 2025.
Chisholm D, Sweeny K, Sheehan P, Rasmussen B, Smit F, Cuijpers P and Saxena S (2016) ‘Scaling-up treatment of depression and anxiety: a global return on investment analysis’, The Lancet Psychiatry, 3(5):415-424.
Deloitte Access Economics (2022) Proposed Occupational Health and Safety Amendment (Psychological Health) Regulations: regulatory impact statement, prepared for WorkSafe Victoria (benefit-cost ratio cited via secondary analyses; the original RIS at engage.vic.gov.au should be confirmed before client-facing reuse).
Dollard MF and Bakker AB (2010) ‘Psychosocial safety climate as a precursor to conducive work environments, psychological health problems, and employee engagement’, Journal of Occupational and Organizational Psychology, 83(3):579-599.
ETUI (European Trade Union Institute) (2025) Latest EU-OSHA survey on workplace risks reveals worrying trends (ESENER 2024 findings), March 2025.
EU-OSHA (European Agency for Safety and Health at Work) (2012) Drivers and barriers for psychosocial risk management: an analysis of the findings of the European Survey of Enterprises on New and Emerging Risks, Publications Office of the European Union, Luxembourg.
Eurocadres (2020) EU health and safety strategy must address psychosocial risks (ESENER-3 findings), November 2020.
HSE (Health and Safety Executive) (2025) Work-related stress, depression or anxiety statistics in Great Britain, 2025, HSE, Bootle.
Jones E, Hyams KC and Wessely S (2003) ‘Screening for vulnerability to psychological disorders in the military: an historical survey’, Journal of Medical Screening, 10(1):40-46.
LaMontagne AD, Keegel T, Louie AM, Ostry A and Landsbergis PA (2007) ‘A systematic review of the job-stress intervention evaluation literature, 1990-2005’, International Journal of Occupational and Environmental Health, 13(3):268-280.
LaMontagne AD, Keegel T, Vallance D, Ostry A and Wolfe R (2008) ‘Job strain-attributable depression in a sample of working Australians: assessing the contribution to health inequalities’, BMC Public Health, 8:181.
LaMontagne AD, Martin A, Page KM, Reavley NJ, Noblet AJ, Milner AJ, Keegel T and Smith PM (2014) ‘Workplace mental health: developing an integrated intervention approach’, BMC Psychiatry, 14:131.
Law Society Journal (2026) Psychosocial safety at work not a ‘soft issue’: new laws as SafeWork NSW claims soar, Law Society of NSW, May 2026.
Madsen IEH, Nyberg ST, Magnusson Hanson LL et al. (IPD-Work Consortium) (2017) ‘Job strain as a risk factor for clinical depression: systematic review and meta-analysis with additional individual participant data’, Psychological Medicine, 47(8):1342-1356.
Marshall RE, Milligan-Saville JS, Mitchell PB, Bryant RA and Harvey SB (2017) ‘A systematic review of the usefulness of pre-employment and pre-duty screening in predicting mental health outcomes amongst emergency workers’, Psychiatry Research, 253:129-137.
Marshall RE, Milligan-Saville JS, Steel Z et al. (2020) ‘A prospective study of pre-employment psychological testing amongst police recruits’, Occupational Medicine, 70(3):162-167.
Niedhammer I, Bertrais S and Witt K (2021) ‘Psychosocial work exposures and health outcomes: a meta-review of 72 literature reviews with meta-analysis’, Scandinavian Journal of Work, Environment and Health, 47(7):489-508.
Niedhammer I, Sultan-Taïeb H, Parent-Thirion A and Chastang J-F (2021) ‘Update of the fractions of cardiovascular diseases and mental disorders attributable to psychosocial work factors in Europe’, International Archives of Occupational and Environmental Health, 95:233-247.
Potter RE, Dollard MF, Owen MS, O’Keeffe V, Bailey T and Leka S (2019) ‘Australian work health and safety policy for the regulation of psychosocial risks: perspectives from key informants’, Policy and Practice in Health and Safety, 17(2):112-132.
Potter RE, Loh MY, Dollard MF, Friebel A, Neser D, Afsharian A, Parker SK and Iles R (2025) ‘Australia’s national laws for worker psychological health: a policy evaluation and psychosocial safety climate analysis’, Safety Science, https://www.sciencedirect.com/science/article/pii/S0925753525001249 (volume and pagination per the published record).
Productivity Commission (2020) Mental health, Inquiry report no. 95, Productivity Commission, Canberra.
Queensland Government (2022) Work Health and Safety (Psychosocial Risks) Amendment Regulation 2022: explanatory notes, Queensland Government, Brisbane.
ReFresh (2026) Case law review: recent prosecutions for psychosocial hazard failures in Australia, January 2026.
Rona RJ, Hooper R, Jones M, Hull L, Browne T, Horn O, Murphy D, Hotopf M and Wessely S (2006) ‘Mental health screening in armed forces before the Iraq war and prevention of subsequent psychological morbidity: follow-up study’, BMJ, 333:991.
Safe Work Australia (2022) Model Code of Practice: managing psychosocial hazards at work, Safe Work Australia, Canberra.
Safe Work Australia (2024) Psychological health and safety in the workplace, Safe Work Australia, Canberra.
Safe Work Australia (2025) Key Work Health and Safety Statistics Australia 2025, Safe Work Australia, Canberra.
SafeWork NSW v Western Sydney Local Health District (No 3) [2025] NSWDC 48.
Squire Patton Boggs (2025) Psychosocial hazards and poor organisational justice: necessary protection or a step too far for employers? (Australia), Employment Law Worldview, November 2025.
Theorell T, Hammarström A, Aronsson G et al. (2015) ‘A systematic review including meta-analysis of work environment and depressive symptoms’, BMC Public Health, 15:738.
Wellbeing Australia (2026) Psychosocial risk at work: governance, legal duties and practical compliance, March 2026.
WHO and ILO (World Health Organization and International Labour Organization) (2022) Mental health at work: policy brief, WHO, Geneva.
This article is general information, not legal advice. Legislative provisions are quoted from current consolidated versions and should be checked against the jurisdiction and date relevant to any specific matter.
Safetysure provides psychosocial risk assessment, gap analysis, and prevention plan development across all Australian jurisdictions. We recommend organisations treat the current regulatory settings as a work design discipline rather than a documentation exercise, and we can assist at every stage of the psychosocial risk management cycle.
Read more on the Code of Practice Changes for NSW 2026