So far as is reasonably practicable: making and recording safety risk decisions in Australian workplaces

A workshop supervisor watches two people lift a large steel sheet onto a press brake. It is heavy, awkward and done dozens of times a day. Everyone knows it is a strain risk. A vacuum lifter would help, but it costs money, takes floor space and slows the first few weeks while people get used to it. Is the business obliged to buy it? Would a second person, a training session and better gloves be enough? And if someone is hurt next month, how would the business show it had done what it should?

Australian work health and safety law does not require businesses to eliminate every risk, which would be impossible, and it does not allow them to choose the cheapest option and hope. It requires them to ensure health and safety so far as is reasonably practicable. That phrase carries a specific meaning: a structured weighing of how likely and how serious harm is, what is known about the hazard and its controls, what controls are available and suitable, and whether their cost is grossly disproportionate to the risk. Many managers use the phrase without knowing how the test works, and fewer record their reasoning in a way that would stand up after an incident.

This article explains the duty and the test, how the hierarchy of control fits, how to distinguish inherent, current and residual risk and verify that controls work before relying on them, how to record decisions, and how to build safety into everyday operations improvement. It is general information, not legal advice. Most Australian jurisdictions have adopted versions of the model Work Health and Safety Act and Regulations; Victoria has its own Occupational Health and Safety Act, which uses a similar “reasonably practicable” concept. Safe Work Australia, state and territory regulators and legal advisers can help with specific situations.

The duty in brief

Under the model work health and safety laws, a person conducting a business or undertaking (PCBU), which includes companies, partnerships, sole traders and many other organisations, has a primary duty to ensure, so far as is reasonably practicable, the health and safety of workers and of other people affected by the work. The duty covers the work environment, plant, structures, systems of work, substances, facilities, information, training, supervision and monitoring.

The law sets an order of preference. A duty holder must eliminate risks to health and safety so far as is reasonably practicable, and if that is not reasonably practicable, minimise them so far as is reasonably practicable.

Other duties sit alongside the primary duty:

• Officers, such as directors and senior managers who make decisions affecting the business, must exercise due diligence to ensure the business complies.
• Workers must take reasonable care for their own health and safety and that of others, and follow reasonable instructions.
• Consultation: businesses must consult workers who are affected by health and safety matters, and must consult, cooperate and coordinate with other duty holders who share the same risks, such as contractors and host employers.

Health includes psychological health. Psychosocial hazards, such as excessive workload, poor support, bullying and harassment, are managed under the same framework.

What “reasonably practicable” means

The model law defines reasonably practicable as what is, or was at a particular time, reasonably able to be done to ensure health and safety, taking into account and weighing up all relevant matters, including:

1. The likelihood of the hazard or risk occurring.
2. The degree of harm that might result.
3. What the person knows, or ought reasonably to know, about the hazard or risk and about ways of eliminating or minimising it.
4. The availability and suitability of ways to eliminate or minimise the risk.
5. After assessing the extent of the risk and the available ways of dealing with it, the cost of doing so, including whether the cost is grossly disproportionate to the risk.

Several practical points follow from this wording.

• The weighing starts with the risk and the controls, not the budget. Cost comes last, after the risk and the available options have been assessed.
• The test is gross disproportion, not a simple cost-benefit balance. A control is not avoided merely because its cost exceeds the expected cost of injuries. It can be set aside on cost grounds only where the cost is grossly disproportionate to the risk. The more severe and likely the harm, the more cost is reasonable.
• “Ought reasonably to know” is objective. It includes what is generally known in the industry, information from regulators, manufacturers’ instructions, standards and approved codes of practice. Courts may refer to an approved code of practice as evidence of what is known about a hazard and its controls. Not having read the guidance is not a defence.
• Availability and suitability matter. A control that exists but does not suit the task, or creates new hazards, may not be reasonably practicable; a well-established control that suits the task usually is.

Safe Work Australia publishes guidance on how to determine what is reasonably practicable, including how cost is weighed. It is worth reading before relying on cost as a reason not to act.

Readers who have worked in the United Kingdom or in industries such as oil and gas may know the similar concept ALARP, “as low as reasonably practicable”. Both rest on the same idea: reduce risk until the cost of further reduction is grossly disproportionate to the benefit.

The hierarchy of control

The model regulations set out a hierarchy of control measures for managing many risks: elimination first; then substitution, isolation and engineering controls; then administrative controls; and personal protective equipment last. Higher-level controls are generally more reliable because they do not depend on people remembering and complying every time. Administrative controls and protective equipment are often needed as well, but rarely justify stopping at that level when higher controls are reasonably practicable. The changing the odds or changing who pays article explains why designing a risk out beats managing it with procedures.

Inherent, current and residual risk

Risk assessments often blur three different states, which leads to optimistic conclusions:

Two disciplines prevent double-counting. First, give credit only for controls that are verified: in place, used as intended and effective. A guard that is routinely removed, a procedure nobody follows or a lifter that is broken is not a control. Second, do not count the same control twice, for example by lowering both likelihood and severity for a control that only affects one.

Verify controls before relying on them

Controls can be grouped by what they do:

• Preventive controls stop the harmful event, such as guarding, interlocks or eliminating a task.
• Detective controls reveal a hazard or developing problem, such as gas detectors, inspections or monitoring.
• Corrective controls limit the harm once an event occurs, such as emergency stops, first aid and spill kits.
• Directive controls tell people what to do, such as procedures, signs and training.

For important risks, ask of each control:

• Is it in place and used, on every shift, not just in the procedure?
• Is it effective: would it actually prevent or limit the harm in the situations that matter?
• Is it independent of the other controls, or would one failure, such as a power outage, a rushed shift or an absent supervisor, defeat several at once?
• Who checks it, how often, and what evidence shows it works?

Layered, independent controls, sometimes called defence in depth, are much stronger than several controls that share a single point of failure. The what engineering failures teach any business article shows how major accidents often follow controls that existed on paper but not in practice.

Record the reasoning, not just the decision

If something goes wrong, a business may need to show what it knew, what it considered and why it decided as it did. A simple record for each significant risk decision:

• The hazard and the people exposed.
• The inherent and current risk, with the evidence used, such as incident history, near misses and observations.
• The control options considered, at each level of the hierarchy.
• Information consulted: codes of practice, guidance, manufacturers’ instructions, specialist advice.
• Workers consulted, and what they said.
• The controls chosen, and why higher-level options were not reasonably practicable, if they were rejected.
• The residual risk expected, and how the controls will be verified.
• Who made the decision, when, and the triggers for review, such as a change in process, an incident or new information.

A short, honest record made at the time is more credible than a long document written after an incident.

Officers’ due diligence

Officers cannot delegate their due diligence duty. Under the model law, due diligence includes taking reasonable steps to:

• Keep up to date with work health and safety matters.
• Understand the operations of the business and the hazards and risks involved.
• Ensure appropriate resources and processes are available and used to eliminate or minimise risks.
• Ensure processes exist for receiving information about incidents, hazards and risks, and responding in a timely way.
• Ensure processes exist for complying with the business’s duties.
• Verify that these resources and processes are provided and used.

In practice, that means directors and senior managers who visit the floor, read incident and near-miss reports, ask about the status of controls for the most serious risks and make sure budgets for safety improvements are considered on their merits. Penalties for breaches can be severe, and several jurisdictions have industrial manslaughter offences.

Build safety into everyday operations improvement

Safety is most effective when it is part of how work is designed and improved, not a separate activity.

• Put safety first in improvement priorities. When choosing which improvement projects to run, rank the severity of safety hazards ahead of other criteria.
• Include ergonomics and safety in every improvement event. A 5S or flow project that moves equipment is an opportunity to remove manual handling, trip hazards and awkward postures.
• Write safety into standard work, at the step where the hazard occurs, not in a separate document.
• Encourage near-miss reporting, and respond visibly to reports.
• Track trends rather than single months. Injury and incident numbers fluctuate. A run chart or control chart of incident rates shows whether a genuine change has occurred, avoiding both panic over one bad month and complacency after one good one.

Many changes that improve flow also improve safety: less handling, less clutter, fewer rushed recoveries from breakdowns and less overtime fatigue.

A worked example

This is an illustrative example. It shows a way of reasoning, not a legal determination. A 25-person fabrication business folds steel sheets on two press brakes. Sheets up to about 30 kg are lifted from a pallet at floor level onto the press brake bed, often by one person, many times a day. Over two years there have been three lower-back strains and several near misses involving sheet edges.

The risk. Likelihood of further musculoskeletal injury is high given the frequency of lifting and the history. Potential harm ranges from strains to serious back injury with long-term effects. Current controls are training in lifting technique, cut-resistant gloves and a rule that sheets over a certain size are lifted by two people, a rule that is often not followed when it is busy.

What is known. The regulator’s code of practice on hazardous manual tasks identifies lifting loads from floor level and repetitive handling as risk factors, and describes mechanical aids and workplace redesign as more effective controls than training. Vacuum lifters and height-adjustable sheet trolleys are widely used in the industry.

Options considered.

• Elimination or substitution: ordering sheets cut to smaller sizes. Possible for some jobs, but many parts need full sheets.
• Engineering: a height-adjustable sheet trolley that presents sheets at bed height, costing a few thousand dollars per press brake; a vacuum lifter on a jib crane, costing about $15,000 for both machines.
• Administrative: enforcing the two-person rule, scheduling and job rotation.
• Protective equipment: gloves, already in use.
• Full automation of sheet loading, costing several hundred thousand dollars.

Weighing. Given the likelihood and severity of harm, the well-known nature of the hazard and the ready availability of suitable aids, the trolleys and vacuum lifter are reasonably practicable; their cost is not grossly disproportionate to the risk. Ordering smaller sheets is adopted where jobs allow. Full automation would further reduce a now much lower residual risk, but its cost is judged grossly disproportionate to that remaining risk at this scale.

Verification and record. The workers who operate the press brakes trial the lifter and trolleys and suggest a change to the trolley height range. A daily pre-use check is added to the lifter, the two-person rule remains for sheets the lifter cannot handle, and the supervisor checks use weekly for the first two months. The decision record lists the options, information consulted, workers’ input, reasons and review triggers. Over the following year, no manual handling injuries are reported at the press brakes, and the near-miss log shows the remaining issues are with sheet edges, which lead to a separate review of edge protection.

Applying this in an Australian business

• Know your duties, including officers’ due diligence and consultation.
• Apply the reasonably practicable test in order: risk, knowledge, available controls, then cost.
• Treat gross disproportion as a high bar, not a simple cost comparison.
• Read the relevant codes of practice and guidance; they define what you ought to know.
• Work down the hierarchy of control before settling for procedures and protective equipment.
• Credit only verified controls when assessing current and residual risk.
• Record your reasoning at the time, including options rejected.
• Consult workers, who know how the work is really done.
• Build safety into improvement and track trends rather than single results.
• Seek advice from regulators, Safe Work Australia guidance or legal advisers for complex situations.

Where safety risk decisions go wrong

• Assuming training and gloves meet the duty when higher controls are available.
• Using cost as the first filter rather than the last.
• Counting controls that exist only on paper.
• Assessing risk once and never revisiting it.
• Not consulting the people who do the work.
• Recording decisions after an incident rather than at the time.
• Treating psychosocial hazards as outside the safety system.

Questions for owners, directors and managers

• Which risks in our business could cause serious harm, and what are the controls for each?
• How do we know those controls are in place and working today?
• For our highest risks, which higher-level controls have we considered and why did we reject any?
• Which codes of practice apply to our work, and have we read them?
• How do we consult workers, and what have they told us recently?
• What do our incident and near-miss trends show over the past two years?
• Could we show, from our records, why our current controls are reasonably practicable?

Bringing it together

“So far as is reasonably practicable” is a structured test, not a slogan. It asks businesses to weigh the likelihood and severity of harm, what they know or ought to know, and which controls are available and suitable, before considering cost, and to set a control aside on cost grounds only when its cost is grossly disproportionate to the risk. Work down the hierarchy of control, give credit only for controls that are verified to work, record the reasoning at the time and consult the people who do the work. Officers must stay engaged and verify. And when safety is built into everyday improvement, the work usually becomes safer and better at the same time.

---

Source: KEVOS editorial notes, drawing on earlier KEVOS handbooks on risk escalation, delegation and ALARP, inherent, current and residual risk, risk controls and defence in depth, and integrated safety and productivity improvement, with reference to the structure of the model Work Health and Safety laws. The worked example is illustrative. This article is general information and does not constitute legal or safety advice.