Study for IOSH Managing Safely by drilling applied decisions, not definitions: practise writing hazard statements that name a source of harm, rating risks by likelihood and severity separately, ordering controls from elimination down to PPE, and explaining why investigating incidents matters. Work through realistic workplace scenarios, compare your reasoning against a self-check rubric, and confirm readiness when you can justify every rating and control choice in one sentence.
Hazard versus risk: the distinction that shapes every answer
A hazard is something with the potential to cause harm; a risk is the chance of that harm, combined with how serious it would be. Every hazard-and-risk question tests whether you keep these two ideas separate.
A workable habit is to phrase hazards as 'source + energy or substance + what it can do'. A trailing cable is a hazard because it can cause a trip; wet flooring near a stair head is a hazard because it can cause a fall down steps; unlabelled decanted solvent is a hazard because it can harm skin and lungs. Risks are then statements about that hazard going wrong: the risk of a delivery worker tripping in a busy corridor, or the risk of dermatitis from repeated splashes.
Practise the swap both directions, because questions do. Given a hazard like a damaged scaffold board, write the risk (a worker falling through or from height, and how badly). Given a risk like 'hand injury from an unguarded blade', name the hazard (the blade at the guillotine). If your sentence contains a probability or a consequence, you wrote a risk; if it names a thing or condition, you wrote a hazard. In the exercise section below, this exact swap is the first check you should run on your own answers.
Rating risk: judge likelihood and severity separately, then combine
Risk rating means estimating how likely the harm is and how severe it would be, then combining both, usually on a matrix. Rating one factor well and ignoring the other is the decision error to train out.
Severity asks: if it goes wrong, what is the worst realistic outcome — first aid, lost time, major injury, or fatality? Likelihood asks: given the task, the people, the environment and the controls already in place, how often is harm credible? A rare task with fatal potential and a frequent task with minor potential can end up with similar overall ratings, which is exactly the comparison to practise.
A common reasoning slip is anchoring on severity: because a fall from height could kill, every height-related item gets the top rating regardless of edge protection, task duration or competence. The better decision is to score severity from the credible consequence, then score likelihood with existing controls in mind, and let the matrix position drive the action. In your practice examples, always note in one short line what pushed likelihood up or down — that justification is the transferable skill, and scenario questions reward visible reasoning over gut ratings.
The hierarchy of control: order the options before you pick one
Controls rank from elimination and reduction, through engineering and administrative measures, to protective equipment. Stronger controls remove or reduce the hazard itself; weaker ones depend on people behaving correctly every time.
Compare two responses to a noisy machine: enclosing it changes the hazard's effect on everyone nearby, every shift, without relying on memory; issuing ear defenders reduces exposure only while they are worn, fitted and maintained. The table below summarises the levels and pairs each with a question that reveals whether you are choosing high enough up the order.
Two pairs are easy to confuse, so name them in practice. Reducing the hazard (a less hazardous substance, a lower voltage) differs from engineering controls (guards, enclosures) — one changes the hazard, the other separates people from it. And administrative controls (procedures, training, signage, supervision) sit near protective equipment at the weaker end because both depend on human compliance. When a scenario offers several fixes, rank every option first, then justify the top feasible one; that ordering habit also exposes distractor answers that are merely plausible.
| Control level | What it does | Workshop example | Ask yourself |
|---|---|---|---|
| Elimination | Removes the hazard entirely | Stop the task requiring work at height by assembling at ground level | Can the task be done without the hazard? |
| Reduction / substitution | Makes the hazard less harmful | Swap a solvent-based cleaner for a water-based one | Is there a less hazardous way or material? |
| Engineering controls | Separate people from the hazard | Fixed guard and interlock on the machine | Can the plant protect people by design? |
| Administrative controls | Shape how people work | Permit-to-work, training, rotation to limit exposure | Does this rely on people following rules? |
| Protective equipment | Protects the individual only | Goggles, gloves, hearing protection | Is this the last line, not the first choice? |
Worked scenario A: the manual handling fix that starts too low
Scenario: warehouse staff repeatedly lift heavy cartons from a low pallet onto a bench, and one worker reports a back strain. The task is to choose and justify controls.
Plausible mistake: the manager orders manual handling training and back-support belts. This is a real decision people make, but both measures are at the weaker end — training and belts depend on every lift being performed correctly, and belts do not reduce the load, the reach, or the repetition. Severity of a single awkward lift is unchanged, so the risk rating barely moves.
Better decision: work up the hierarchy first. Elimination or reduction: can the supplier deliver directly to the bench, or on smaller cartons? Engineering: a scissor lift table or pallet elevator removes the low-level lifting posture. Only after those are infeasible do procedures and training enter. Why it matters: the same likelihood and severity scores that justified attention now justify a stronger control, and the reasoning — remove the hazard before coaching people around it — is exactly the pattern scenario questions ask you to demonstrate. Write the rejected options and your reason; that sentence is your exam-ready justification.
Worked scenario B: decanting solvent and mislabelling the hazard
Scenario: a maintenance worker decants solvent from a drum into an unlabelled bottle in a poorly ventilated store. Hazards, risks and controls must all be identified.
Plausible mistake: the learner writes 'spillage' as the hazard and 'clean it up quickly' as the control. Spillage is an event, not a source of harm; the hazards are the flammable liquid and its vapour, and the risk is fire plus inhalation and skin harm. With that reframing, controls change: an unlabelled bottle and poor ventilation become central, whereas 'clean quickly' only addresses one accident path.
Better decision: state the hazards (solvent liquid and vapour), rate the risk for fire and for health effects separately, then order controls — buy ready-diluted solution or decant only what is needed (reduction/elimination of decanting), fit a dispensing pump and provide labelled containers and ventilation (engineering and reduction), keep quantities small with storage rules (administrative), and gloves and eye protection last. Why it matters: mislabelling the hazard steers every later answer toward tidiness instead of the actual harm, and exam scenarios are built so that an imprecise opening statement cascades. Check scenario B's own structure: one sentence for the hazard, one for the risk, then the ordered controls.
Investigation, responsibility and measuring safety: link the ideas, don't list them
The course connects why incidents are investigated, who holds responsibility for safety, and how performance is measured. Treating these as three memorised lists hides the logic that ties them together.
Investigation serves prevention and learning: find the immediate causes (the event and the direct agent), then the underlying causes — the working conditions, organisation and decisions that allowed it. Trace the scenario B spill-forward if a fire had occurred: the unlabelled bottle is immediate, but no decanting procedure and no supervision are underlying. Practising the two-layer chain prevents answers that stop at 'worker was careless', which explains nothing and prevents nothing.
Responsibility runs through the organisation rather than sitting only with specialists: directors and managers set direction and provide resources, line managers supervise and enforce, and workers follow safe systems and report problems. Measurement then closes the loop: proactive checks (inspections, near-miss reporting, monitoring conditions) find problems before harm, while reactive data (incidents, ill-health cases) shows what has already gone wrong. When you revise, rehearse the link in one sentence each: investigate to find causes, assign responsibility to fix them, measure to confirm the fix works. That chain is a natural exam-structure template.
A preparation sequence with a self-check rubric and readiness checks
Prepare in four passes: vocabulary precision, rating drills, control-ordering drills, then full scenario writing. Close with a rubric-based self-review and defined readiness checks rather than rereading notes.
Adaptable sequence: pass one (days 1–2), rewrite ten hazards from your own workplace as 'source + energy/substance + harm' statements and swap each into risk form. Pass two (days 3–4), rate eight tasks before and after controls on a simple grid, noting one-line justifications. Pass three (days 5–6), take five real fixes you know of and rank every option by hierarchy before revealing what was actually done. Pass four (final days), write three complete scenario answers — hazard, risk, rating, ordered controls, who is responsible — under time pressure, then score them with the rubric below. Keep sessions short and grounded in tasks you can observe; familiarity makes the vocabulary stick.
Practical exercise with expected observations: pick one ordinary task, such as changing a ceiling-mounted projector bulb. Write the hazard (work at height from a ladder, electrical energy), the risk (fall and electric shock, severity high, likelihood low with a stable ladder and isolation), then order controls: eliminate by using a pole-mounted fitting or a access platform, engineering by isolating the circuit, administrative by a two-person procedure, eye protection last. Expected observations: your likelihood score drops when controls remove exposure, not merely because you want a lower number; severity rarely changes unless the hazard itself is reduced. Self-check rubric — score each of your practice answers: 1) hazard named as a source of harm, yes/no; 2) risk includes both likelihood and severity, yes/no; 3) at least one option above administrative control considered, yes/no; 4) rejection reasons stated in one sentence, yes/no. Readiness checks: you can produce a compliant answer for three unseen tasks without notes; you can rank any five-option control list in under a minute; you can explain the investigation chain (immediate to underlying causes) for one incident you know. Treat rubric outcomes as learning milestones, not predictions of any result; for administrative details of the qualification, rely on the issuer's own pages.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.