Energy-based safety is enjoying a recent revival in health and safety circles across Austraia and New Zealand. The concept is certainly attractive because it is simple and supposes that harm occurs when energy is transferred to a person in amounts the body cannot absorb. Find the energy and the hazard is found Subsequently, control the energy and the risk is controlled.
Like most simple ideas in work safety, it has merit, it is genuinely useful, and a little incomplete. This article explains where the energy lens came from, what the evidence says it does well, and where an organisation that relies on it exclusively will develop blind spots. Our view, developed across audits, investigations is that energy-based safety earns a place in most hazard management systems but it should be treated as one lens among several.
The origins of energy based safety
The energy lens for assessing hazards is not new. William Haddon Jr, a physician appointed in 1966 as the first administrator of the agency that became the United States National Highway Traffic Safety Administration, formalised it more than fifty years ago. Haddon argued that injury is fundamentally an ecological event – a transfer of mechanical, thermal, chemical, electrical, or radiant energy at rates and in amounts that damage living tissue (Haddon 1970; Haddon 1973).
From that single premise, Haddon derived ten countermeasure strategies, arranged in logical sequence. The first six act before contact – prevent the energy from being marshalled at all, reduce its amount, prevent its release, modify the rate or distribution of that release, separate the energy from people in space or time, and place a barrier between them. The remaining four address the moment of contact and its aftermath – modify contact surfaces, strengthen the person or structure exposed, detect and limit damage quickly, and stabilise and rehabilitate (Haddon 1973).
Our readers familiar with Australian WHS law will recognise the family resemblance. Section 17 of the WHS Act, mirrored in regulation 35, requires duty holders to eliminate risks so far as is reasonably practicable, and to minimise them only where elimination is not. Regulation 36 then sets the order for minimising what remains: substitution, isolation, and engineering controls first, then administrative controls, then personal protective equipment. That statutory sequence is essentially Haddon’s logic in legal form. Controls that act directly on the energy rank above controls that depend on people behaving as intended. Energy-based safety did not invent this ordering; it explains why the ordering exists.
The modern revival of energy based safety
The current wave of interest owes most to Professor Matthew Hallowell and the Construction Safety Research Alliance at the University of Colorado. The research programme addressed an uncomfortable pattern which showed that across two decades, recordable injury rates in construction and utilities fell substantially while serious injury and fatality (SIF) rates barely moved. The hazards that produce sprains and lacerations, it turns out, are not the same hazards that kill people. Surprise surprise!! Systems tuned to reduce total recordable injuries can look excellent on paper while doing little about fatal risk. We have written about the governance version of that trap in The numbers were always green and about the underlying measurement problem in avoiding the watermelon effect.
Hallowell’s team also quantified something field practitioners have long suspected that hazard recognition is far weaker than assumed by many in the health and safety profession. Across thousands of hours of structured observation, workers in pre-job briefings identified fewer than half of the hazards actually present in their work, and the blind spots were consistent across trades, experience levels, and education. Gravity and motion hazards were recognised almost instinctively. Mechanical, pressure, and temperature hazards were routinely missed (Hallowell 2021).
The energy wheel was developed as a corrective concept in that it organises hazards into ten energy sources: gravity, motion, mechanical, electrical, pressure, sound, temperature, chemical, radiation, and biological. Used as a prompt during pre-job briefings and workplace inspections, it directs attention to the categories the brain skips. In field experiments, the tool improved hazard recognition by roughly 30 per cent (Hallowell 2021). Later work added high-energy hazard icons, a severity classification model, and high energy control assessments. The last of these measures the proportion of high-energy hazards protected by a direct control, meaning a control that targets the energy itself and holds even when someone makes a mistake (Hallowell 2026).
What an energy based safety review looks like
An energy-based review of a task starts with a different question. Instead of asking what could go wrong, which invites traditional speculation, the reviewer asks what energy is present, how much, and what stands between it and a person. A suspended load is gravitational energy, a reversing forklift is kinetic energy carried by several tonnes of machine, a point made starkly by the fatality we examined in early lessons from a fatal forklift rollover. A pressurised hydraulic line, a charged capacitor bank, a silo of stored grain, and a furnace at operating temperature are all reservoirs of energy waiting for a release path.
The follow-up question is the discipline: for each high-energy source, is there a direct control? A trench shield is a direct control for soil collapse. A machine guard is a direct control for entanglement. A hard hat is not a direct control for a falling scaffold clamp; it is a last line that may or may not hold. That distinction maps cleanly onto the hierarchy of control and gives supervisors a fast, defensible way to test whether a safe work method statement or permit to work is describing real protection or administrative comfort.
The case for the energy safety lens
Four strengths stand out from the evidence and from our own field experience.
It is concrete and teachable
Energy is physical and observable. A worker does not need a diploma to see that a load is suspended overhead or that a line is under pressure. The vocabulary transfers across trades and across language barriers, which matters on Australian sites where crews change weekly. The measured improvement in hazard recognition is one of the better-evidenced results in safety training research (Hallowell 2021).
It discriminates severity
Traditional risk matrices ask assessors to estimate consequence, and assessors are demonstrably poor at it. Energy magnitude is a physical proxy for the worst credible outcome. A task involving 11 kV switchgear or a nine-tonne counterweight simply cannot produce a trivial worst case, whatever the likelihood estimate says. This gives organisations a rational basis for concentrating effort on fatal risk rather than spreading attention evenly across every entry in the risk register.
It privileges engineering over behaviour
Because the lens focuses on the energy rather than the person, it naturally pushes solutions up the hierarchy of control. That is a welcome counterweight to decades of behaviour-based programs that asked workers to be more careful around unchanged hazards. As we explored in why people take risks at work, risk-taking is usually a rational adaptation to the system people work in, and lecturing rarely changes it. Removing or isolating the energy does.
It improves what gets measured
Counting the percentage of high-energy tasks that have a verified direct control is a leading indicator with a direct causal link to fatality prevention. It is considerably harder to game than a lost time injury frequency rate, and it gives boards and officers something meaningful to ask about when exercising due diligence.
The limits of the energy safety lens
The weaknesses are just as real, and an organisation adopting energy-based safety should walk in with its eyes open.
Chronic health harm fits awkwardly in the model
The energy model was built for acute injury: rapid, high-magnitude transfers. Occupational disease works differently. Respirable crystalline silica is technically a chemical hazard on the wheel, but the harm accumulates invisibly across thousands of low-dose exposures over years, and no pre-job energy scan will reveal it. The same is true of noise-induced hearing loss, diesel particulate, and welding fume. These hazards demand exposure measurement, health monitoring, and long-horizon controls of the kind delivered through structured occupational hygiene and silica monitoring, not a briefing prompt. The scale of that blind spot is easy to miss because the national figures obscure it. Safe Work Australia’s headline count of 188 worker deaths in 2024 is a traumatic injury dataset, and it expressly excludes deaths from occupational disease, which are not captured in any comparable national series. Meanwhile diseases and conditions have grown from about a quarter of serious workers’ compensation claims a decade ago to more than a third. An organisation whose entire hazard vocabulary is energy will systematically underweight harm that the headline fatality number was never designed to count.
Psychosocial hazards have no energy source
There is no meaningful sense in which excessive job demands, poor organisational justice, or workplace sexual harassment are energy transfers. Yet these are now regulated hazards under the psychosocial provisions embedded in Australian WHS regulations, and the evidence for their health consequences is substantial, as we set out in regulating psychological health at work. A safety system organised purely around the energy wheel has no shelf to put these hazards on, and what has no shelf tends not to get managed.
Energy is the agent for injury but it’s not the explanation
Energy-based analysis is superb at describing the final second of an incident and largely silent on the preceding five years. The uncontrolled energy is the proximate cause; the reasons the control was absent, degraded, or bypassed are organisational. Budget pressure, production incentives, normalised deviance, inadequate supervision, and paper systems that diverge from work as done are where fatal incidents are actually incubated. Australian inquiries into major industrial disasters have repeatedly made some version of this point. An organisation can achieve impressive energy-control statistics while its assurance processes quietly decay, which is why the lens needs pairing with genuine verification of controls in the field rather than in the document register.
It can become another relabelling exercise
We have consistently reviewed risk registers where every hazard was dutifully re-sorted into energy categories and not a single control changed. The lens only earns its value when it alters decisions at the work face i.e when a high-energy task with no direct control triggers redesign rather than a toolbox talk. Adopted as vocabulary rather than discipline, energy-based safety adds clutter to systems already carrying too much of it.
Terminology can collide with Australian law
The direct and alternative control language used in the North American literature is useful shorthand but it is not the legal test. The statutory test at section 17 of the WHS Act and regulation 35 is to eliminate risks so far as is reasonably practicable and otherwise to minimise them, applying regulation 36 to what remains. A PPE-dependent control might be classified as merely alternative in energy-based terms yet still be a required layer legally. Organisations should map the two frameworks explicitly rather than letting imported terminology drift into procedures and, eventually, into court documents.
Using the energy safety lens
Our practical advice is to use the energy wheel where the evidence supports it. Things like pre-job briefings, workplace inspections, hazard identification for high-risk construction work, and design reviews. Use energy magnitude to prioritise which risks receive engineering attention and which controls receive field verification. Adopt the direct-control question as a standing test of whether documented controls, including PPE that workers may not actually wear, would survive a human error.
Then deliberately run the other lenses across your hazards. Keep an exposure-based programme for chronic health hazards, with monitoring data rather than assumptions. Keep a psychosocial risk process that meets the regulatory requirements now in force. Keep asking the organisational questions that energy analysis cannot answer whether controls verified on paper exist on the ground, and whether anyone would know if they did not.
Energy-based safety is a genuinely valuable addition to Australian work safety practice. It is also, on its own, a partial and incomplete map of hazards. The organisations that benefit most treat it as Haddon intended: a sharp analytical tool inside a broader system of thought not a substitute for one.
Safetysure helps organisations integrate energy-based methods into their hazard identification, auditing, and assurance programmes, and independently verifies whether critical controls exist in practice. To discuss how the approach could work in a particular operation, the team can be reached through the contact page or on 1300 087 888. Safetysure is an ISO 9001/45001/14001-accredited workplace health, safety and occupational hygiene consultancy based in Brisbane. This article is general information and not legal advice.
References
Haddon, W Jr (1970) ‘On the escape of tigers: an ecologic note’, American Journal of Public Health, 60(12):2229–2234.
Haddon, W Jr (1973) ‘Energy damage and the ten countermeasure strategies’, Journal of Trauma, 13(4):321–331. The same paper was also published as Human Factors, 15(4):355–366.
Hallowell, MR (2021) ‘The energy wheel: the art and science of energy-based hazard recognition’, Professional Safety, 66(12):27–33.
Hallowell, MR (2026) Energy-based safety: a scientific approach to preventing serious injuries and fatalities (SIFs), Routledge.
Safe Work Australia (2025) Key work health and safety statistics Australia 2025, Safe Work Australia, Canberra.