SHARE Knowledge Hub

Reactive Monitoring

Reactive monitoring learns from workplace accidents, near misses, ill health and failed controls after they occur. It traces the harmful energy or exposure, the people affected and the safeguards that did not work, including events where chance prevented injury. Protect people first, decide the correct GB or Northern Ireland reporting route, investigate causes and verify stronger controls before treating the case as closed.

Updated 2026-10-19Great Britain and Northern IrelandVersion 2.2

Understanding Reactive Monitoring

Reactive monitoring learns from workplace accidents, near misses, ill health and failed controls after they occur. It traces the harmful energy or exposure, the people affected and the safeguards that did not work, including events where chance prevented injury. Protect people first, decide the correct GB or Northern Ireland reporting route, investigate causes and verify stronger controls before treating the case as closed.

Why this matters at work

The evidence set should include event narrative, credible worst consequence, barrier status, potential-severity rationale, peer challenge and escalation decision. Unsafe shortcut: ranking a dropped load that narrowly missed people as trivial because no injury occurred. Escalate when potential fatality, major accident, multiple casualty, irreversible disease or major environmental harm was credibly possible. Where evidence is incomplete, keep the conclusion conditional and control the residual risk rather than forcing certainty.

Set investigation effort using both what happened and what could credibly have happened. Base potential severity on the actual energy or exposure, people plausibly present, failed and surviving safeguards, and one or more realistic changes in conditions. Avoid giving every small deviation an imaginary catastrophic outcome, but do not classify a narrowly missed fatal exposure as trivial merely because nobody was hurt. Repeated events and incomplete evidence can also justify deeper investigation. Record the event narrative, scenario, barrier state, affected population, uncertainty and reasons for the investigation level. Obtain suitable challenge where the potential consequence is serious or disputed. Make safe the current work while the evidence is gathered and check external reporting separately. The selected scope should allow the team to examine immediate, underlying and organisational causes relevant to the failure. If later evidence changes the credible potential, revise the classification and investigation rather than preserving an early low score to meet a closure target.

How organisations manage this in practice

The evidence set should include a controlled scope statement, responsibility map, event taxonomy, escalation matrix, legal register, data dictionary and review cycle. Unsafe shortcut: treating reactive monitoring as an accident-counting exercise or as a substitute for preventive risk control. Escalate when a serious event, high-potential near miss, emerging health pattern or critical-control failure falls outside the reporting system. A competent reviewer should be able to reconstruct the event, evidence, causal reasoning, selected controls, action closure and effectiveness decision.

The evidence set should include risk-control map, leading and lagging indicators, inspection and test results, event data, thresholds and management review records. Unsafe shortcut: presenting a low injury count as proof of effective control without checking exposure, underreporting or critical-control performance. Escalate when lagging data improve while active checks, assurance coverage, worker reports or barrier condition deteriorate. Closure requires evidence that the control exists and functions; completion of a task or issue of a procedure is not enough.

An investigation report is passive learning until its findings change practice and are checked. Preserve the lesson’s mechanism, affected equipment or task, accountable recipient, required change and verification. Review handovers, contractor interfaces and personnel changes so the organisation retains the lesson after the original investigators leave. Source basis: HSE learning organisations.

One near miss can create parallel duties

Immediate protection, statutory reporting where required, and investigation can need action together. Reporting must not wait simply because causal analysis is unfinished.

  1. Protect people now

    Make the situation safe through the competent site arrangements and preserve relevant evidence without creating further exposure.

  2. Check reporting independently

    The responsible person checks the applicable GB or NI categories, thresholds and deadlines from current official guidance.

  3. Investigate and improve

    Test evidence, select stronger controls and name an owner and acceptance test; a low injury count does not prove control health.

  4. Verify and share

    Keep the action open until controls work in real use, then return learning to similar tasks. Numbered signal types are categories, not incident counts.

Read the complete task-specific control examples

ERICPD is a planning cross-map: Discipline verifies and sustains every control. ESEAP compares effectiveness. A label does not authorise the task; retain the owner, evidence and stop or change decision.

Lifting pallets after a dropped-load near miss · ERICPD

Eliminate
Remove the need for the suspended-load movement by changing delivery or stock layout where feasible.
Reduce
Use a suitable alternative load unit or handling method that prevents the identified instability after an assessed trial.
Isolate
Keep people outside the lifting/drop zone and secure the affected equipment and damaged load against reuse.
Control
Provide suitable rated lifting equipment, load restraint and stable support; investigate attachment and pallet condition.
PPE
Use task-assessed head, hand and foot protection; PPE cannot protect a person underneath a falling load.
Discipline
Verify equipment repairs and the revised load-handling method under controlled conditions before closing the finding.

Lifting pallets after a dropped-load near miss · ESEAP

Elimination
Remove the need for the suspended-load movement by changing delivery or stock layout where feasible.
Substitution
No separate approved substitution is specified; assess a genuinely safer material or process before relying on lower controls.
Engineering controls
Use a suitable alternative load unit or handling method that prevents the identified instability after an assessed trial.
Administrative controls
Reassess the lift plan, load information, competence and communication using evidence from the near miss.
PPE
Use task-assessed head, hand and foot protection; PPE cannot protect a person underneath a falling load.

Reviewing powder use after breathing complaints · ERICPD

Eliminate
Remove or suspend the changed powder process where feasible and restore a suitable assessed alternative.
Reduce
Trial a suitable lower-hazard formulation or non-dusty form after reviewing respiratory, skin, fire and product-quality risks.
Isolate
Separate the affected process from nearby workers and prevent contaminated dust spreading through the workplace.
Control
Verify enclosure, extraction and cleaning against the actual new material and rate of use.
PPE
Select adequate suitable RPE and other protection for remaining exposure; PPE does not replace fixing failed source control.
Discipline
Check corrective action in normal work, review further health reports and retain the authorised restart evidence.

Reviewing powder use after breathing complaints · ESEAP

Elimination
Remove or suspend the changed powder process where feasible and restore a suitable assessed alternative.
Substitution
Trial a suitable lower-hazard formulation or non-dusty form after reviewing respiratory, skin, fire and product-quality risks.
Engineering controls
Separate the affected process from nearby workers and prevent contaminated dust spreading through the workplace.
Administrative controls
Investigate the change and exposure route, involve workers and occupational health, and agree evidence needed before restarting.
PPE
Select adequate suitable RPE and other protection for remaining exposure; PPE does not replace fixing failed source control.

What is reactive monitoring?

Reactive monitoring collects and acts on evidence of accidents, near misses, ill health, damage and control failures after they occur. Its purpose is to understand how risk escaped control and prevent recurrence.

A report is the start of a chain: protect people, capture facts, assess actual and credible potential harm, decide whether the law requires notification, investigate proportionately, select stronger controls and verify they work. A low injury count can coexist with high potential severity, underreporting or weak safeguards. Active monitoring tests controls before harm; both views are needed.

Reactive monitoring itself is not the physical hazard. The underlying energy, substance, work organisation or failed barrier harms people. The examples below connect events to those mechanisms; a near miss is valuable because chance may be the only reason nobody was injured.

What harm can a reactive signal reveal?
Signal and exposure routeActual or credible harmImmediate control question
A dropped load lands near a worker in a construction area.The load’s energy could cause crush injury or death even when this event injures nobody.Is lifting stopped, the area isolated and the load path and exclusion control verified? HSE lifting operations
A vehicle crosses a pedestrian route in a warehouse.A collision could cause crushing, fracture or fatal injury.Are vehicles and people separated while the route and supervision are examined? HSE pedestrian/vehicle separation
Repeated wet-floor falls or near falls in hospitality.Slip energy can cause sprains, fractures or head injury; a similar event may affect a visitor.Is the source removed or contained before reopening the route? HSE slip prevention
Several workers report breathing symptoms after a process change.Airborne exposure may cause acute irritation or longer-term disease even without one accident event.Are exposure controls checked and suitable health advice sought while information stays confidential? HSE COSHH assessment

How can the hierarchy of controls turn incident and ill-health findings into effective risk reduction?

Move from the event to a tested explanation, a stronger control and evidence that the control works in everyday work.

Start with care and scene safety. Preserve physical and digital evidence without putting anyone in danger. Separate direct observation, witness account, inference and unknown facts; invite workers and contractors to explain actual task conditions. HSE HSG245 sets out gathering information, analysing it, identifying risk controls, and making and implementing an action plan.

Look beyond the final action. A person walked into a route, but why was the route open, the barrier missing, the handover unclear or the vehicle schedule changed? Compare plausible explanations and seek disconfirming evidence. Assign an owner to each interim restriction and final correction. Close an action only after testing the installed control and checking whether it remains effective.

ERICPD (Eliminate, Reduce including substitution, Isolate, Control, Personal Protective Equipment, Discipline) is a useful six-stage control-planning framework, not a legal reporting category. Apply it to the underlying hazard found by the event. The strongest practicable control should address the actual mechanism; PPE and discipline cannot compensate for a known missing physical barrier.

Follow one illustrative near miss: a pedestrian steps through a missing gate into a forklift path and a driver brakes. Record the event, make the scene safe, and trace the missing gate and traffic design before choosing corrective action. ERICPD supports the action discussion; the HSE five-level hierarchy identifies whether a measure removes, substitutes, engineers or merely administers the hazard. Reporting compliance and a new briefing cannot replace a feasible physical correction. HSE control hierarchy.

On a narrow screen, scroll sideways for the source-control choice, verification and stop point.

How do controls change for breathing-symptom cluster after a material change?

This second pathway uses the same ERICPD prompt and the HSE five-level hierarchy. Each proposed control needs a competent feasibility decision and proof at the actual site. Official task guidance supports the method; the example does not authorise a particular installation.

On a narrow screen, scroll sideways for feasibility, proof and stop decisions.

How does reactive monitoring apply across sectors?

The reporting and learning chain is common, but the harmful agent, exposed people and effective control differ by task.

These are illustrative events, not claims that every site in a sector has the same risk. Triage should consider both the observed result and what could credibly have happened if a person or safeguard had been in a slightly different position. A common dashboard can compare themes; the corrective control must remain specific to the mechanism.

How the same learning chain transfers between sectors
Sector, signal and peoplePotential harm and failed controlImmediate and longer-term decision
Construction: suspended material falls near a contractor.Crush or fatal injury; exclusion zone or lifting plan may have failed.Stop lifting, keep people clear, preserve evidence and review the lifting control before restart. HSE lifting operations
Logistics: forklift enters a pedestrian crossing.Vehicle strike; physical segregation, route design or shift coordination may be weak.Separate traffic now, examine task and layout, then verify revised routes in live shifts. HSE pedestrian/vehicle separation
Hospitality: repeated slips in a service corridor.Fracture or head injury to workers or visitors; drainage, floor condition or spill response may be inadequate.Control the route, investigate source, and verify a durable floor or drainage correction. HSE slip prevention
Manufacturing: several workers report breathing symptoms after material change.Inhalation exposure may cause illness; ventilation or change control may have failed.Check exposure controls and obtain suitable health advice, protect private records and review the change before continued use. HSE COSHH assessment

What do a high-potential near miss and a health pattern teach?

An event without injury may still expose a fatal pathway, while occupational ill health may appear as several weak signals rather than one accident.

A dropped load with no injury but high potential

Task, material and setting: At an illustrative construction site, a suspended package drops into a temporarily open walkway. Nobody is struck. The project has a lifting plan and a shared contractor access route.

People and prerequisites: The lifting supervisor, contractor workforce, site manager and any visitors near the route may be affected. Before work resumes, the responsible team needs the lifting equipment status, plan, exclusion arrangements and witness accounts.

Before: Stop the lifting activity, protect anyone at risk, keep the area controlled and preserve available evidence without compromising immediate safety. Record who was present and who controls the scene.

During: Triage the event by credible crush potential and possible statutory dangerous-occurrence criteria, not by zero actual injury. A competent investigator reconstructs the sequence, load path, barriers, supervision, contractor interface and equipment condition.

Hold or stop: Keep the lifting operation on hold if the load path, equipment, exclusion boundary or cause is uncertain. Immediate danger triggers the site emergency arrangements. The responsible person checks GB or NI notification using the applicable law; do not assume every dropped object is automatically reportable.

Completion and hand back: The authorised site owner accepts a revised lifting and access arrangement only after required equipment and boundary checks, worker briefing and confirmation that interim actions are closed or controlled.

Evidence: Retain facts, images or logs, witness accounts, statutory-reporting decision, causal analysis, selected higher-order controls, owner, verification and later effectiveness review.

Control source: HSE lifting operations. The site details are illustrative; verify the actual task and jurisdiction before release.

The construction example remains a high-potential learning event even if no statutory report is ultimately required. Reportability and investigation depth are distinct decisions.

Breathing symptoms after a material change

Task, material and setting: At an illustrative manufacturing site, three workers separately report breathing discomfort after a new material grade is introduced. There is no single dramatic spill or injury event.

People and prerequisites: Affected workers need support and suitable medical or occupational-health advice. The process owner, supervisor and competent exposure assessor need the material change record, work pattern and control information; clinical details stay with authorised people.

Before: Record each report respectfully, check whether exposure may be continuing, and apply proportionate interim restrictions or controls. Seek competent health and exposure advice rather than waiting for a statistical trend.

During: Examine material identity, ventilation performance, tasks, timing and comparable reports. Keep symptoms and private records separate from a widely shared incident narrative. The responsible person checks whether a diagnosed reportable disease and specified exposure criteria are met under the relevant GB or NI rules.

Hold or stop: If control performance is unknown or symptoms continue, restrict the affected process and escalate. Acute distress or suspected serious exposure follows medical and emergency arrangements. Do not conclude that the material is safe merely because no report under the Reporting of Injuries, Diseases and Dangerous Occurrences Regulations (RIDDOR) is yet required.

Completion and hand back: The process owner verifies the revised material or exposure control and obtains competent advice before normal use. Workers receive a clear explanation of controls without disclosure of someone else’s health information.

Evidence: Retain the change and exposure-control records, anonymised trend, assessment, medical advice held with lawful access, statutory-reporting decision, action test and effectiveness review.

Control source: HSE COSHH assessment. The site details are illustrative; verify the actual task and jurisdiction before release.

These worked examples provide decision logic only. Actual site restrictions, medical action, statutory classification and restart approval require the applicable procedures and competent judgement.

What should a reactive monitoring dashboard show?

Show potential harm, barrier condition, reporting coverage and action effectiveness alongside incident counts and rates.

A low rate may mean few exposures, good controls, chance, underreporting or a small denominator. Record the definition and population behind each measure and explain changes in reporting routes or classification. Avoid safety triangles as a prediction that a fixed number of minor events leads to one serious accident.

A monthly count should lead to a control decision. If forklift near misses fall while pedestrian-route defects remain open, the manager should not infer that separation has improved; they should walk the route, check the physical barrier and review whether people still report weak signals. Record the owner and closure evidence for each critical correction, then sample the route again after work patterns change. HSE process-safety indicator guidance explains why outcome counts need to be read alongside control performance.

Signals a dashboard should show
MeasureInterpretation check
Actual and potential severitySeparate outcome from credible potential; show high-potential near misses and critical-control failures.
Report coverage and timelinessInclude workers, contractors, sites and health reports; a rise may reflect improved trust rather than worsening control.
Action and verificationShow interim restrictions, overdue critical actions, tests completed and effectiveness results, not only closure percentages.
Exposure and denominatorState work hours, tasks, people or relevant opportunity where a rate is used; flag small numbers and changed definitions.
Active control evidenceCompare event trends with inspections, maintenance, exposure checks and worker observations to challenge false assurance.

HSE HSG254 helps major-hazard sites develop process-safety indicators; its approach is useful for critical controls but does not create a universal target for every business. Protect individual health and witness data when sharing trend information.

What does a strong incident-to-learning pathway look like?

Each hand-off needs an owner and evidence, from immediate protection and legal classification through causal learning and a tested corrective control.

An organisation should define a route for workers, contractors and the public to report relevant events. It should keep statutory reporting distinct from the wider internal learning system. The duty manager may control immediate danger while another responsible person decides RIDDOR and a competent investigator scopes the causal work.

From event to verified learning
StepQuestion and evidence
Protect and captureIs anyone in danger? Provide care, control the scene and capture prompt facts from all affected work groups.
Classify and notifyWhat happened, what could credibly have happened, and which GB/NI statutory or sector route applies? Record the decision and owner.
InvestigateWhat sequence, failed barriers and work conditions explain the event? Separate fact from inference and test alternatives.
Choose and actWhich higher-order control addresses the mechanism? Assign resources, interim restrictions, owners and acceptance criteria.
Verify and learnHas the change been installed and tested in real work? Check recurrence, transfer learning and update risk controls and management review.

Closing a task in software is not the same as proving that the hazard is controlled. Revisit the action after enough real work has occurred to see whether the new barrier survives normal shifts, maintenance and contractor interfaces.

What separates a weak investigation from a strong one?

A strong investigation reconstructs the event and tests how multiple controls and work conditions shaped it; a weak one stops at a blame label or first visible error.

“Worker failed to follow procedure” is a hypothesis, not a complete explanation. Ask whether the procedure matched the job, whether equipment and time were available, what supervision and handover conveyed, and why the barrier was vulnerable. Interview people without blame, document uncertainty and allow a competent reviewer to challenge causal statements.

The corrective measure must follow the causal finding. If the route is poorly designed, retraining alone leaves the collision risk. If a critical guard repeatedly fails, a reminder poster does not repair it. Record rejected options, interim restrictions, owners, commissioning evidence and the test that will show the new control works.

How should reactive events be triaged?

Triage immediate safety, potential severity, legal reporting, failed barriers and the investigation needed before routine work resumes.

The first report may be incomplete. Keep classification open where injury, diagnosis, equipment failure or sector scope is unclear; revise it when facts improve and preserve the audit trail. Set prompt escalation for high-potential events and critical-control failures even if the statutory test is not met. Do not use one simple injury score to decide every case.

The first triage separates immediate danger from learning and legal reporting. Make the area safe and preserve the facts; a competent person then checks the actual event against the applicable GB or NI reporting route while an investigator tests failed controls. Do not wait for a polished root-cause report before making a time-sensitive reporting decision, and do not treat a submitted report as evidence that the control is restored. See HSE reporting timing and HSENI incident reporting.

Decisions to make at the first review
QuestionAction and record
Is anyone in immediate danger or in need of care?Protect people and call the relevant emergency or medical route. Preserve evidence only when safe.
What is actual and credible potential severity?Describe the energy or exposure and the failed barriers. Escalate high potential even when actual injury is minor or absent.
Could statutory or sector notification apply?Use the GB or NI legal route and responsible person; record the reasoning, notification and timing. Do not equate every near miss with a dangerous occurrence.
What interim restriction is needed?Name the owner, affected task, condition for release and communication to workers and contractors.
What investigation is proportionate?Set competence, independence, scope, evidence plan and worker involvement based on the actual and potential consequences.

Illustrative reporting triage record

Run immediate protection, legal reporting and causal investigation in parallel. The responsible person records jurisdiction, event time, incapacity or diagnosis information, category under review, next decision time, the current regulator source and submission reference. An investigation deadline does not extend a reporting deadline.

GB incapacity example: record the accident time as day 0 and check whether normal-work incapacity exceeds seven consecutive days; exclude the accident day and include non-working days. Where this category applies, notify within 15 days of the accident. NI example: separately check the over-three-day category and submit within 10 days. These are illustrative incapacity routes, not universal reporting periods. Qualifying fatalities, specified/major injuries and dangerous occurrences need prompt notification under their applicable route. The reporting owner checks the category without waiting for causal closure, retains confirmation, and rechecks if the facts change.

HSENI reporting criteria and timing; HSE reporting timing.

Where does reactive monitoring commonly fail?

The system fails when people cannot report, serious potential is downgraded, cases are coded inconsistently, actions are weak or closure is accepted without a working control.

Underreporting can arise from difficult forms, retaliation fears, contractor exclusion or lack of feedback. Blame can suppress evidence about workload and design. A dashboard may hide worsening risks if denominators, case definitions, exposure, illness data or critical-barrier tests are missing. A due date is an escalation signal, not proof that an action has reduced risk.

Escalate repeated near misses, occupational-health patterns, disabled safeguards and long-open critical actions. Investigate the reporting system itself if major events appear only after outside complaints or regulator contact. Return the learning to workers and check that the corrective measure works on real shifts.

What evidence should reactive monitoring hold?

A reviewer should be able to trace a report from the original event to legal triage, causal reasoning, a tested control and a later effectiveness decision.

Retain enough detail to reconstruct what happened without circulating personal medical information unnecessarily. Link duplicate reports while preserving what each witness actually said. Record version changes to classifications, causal conclusions and actions rather than overwriting history.

Records that make learning reviewable
RecordWhat it should prove
Initial event and care recordEvent identity, people, task, actual outcome, immediate control and support, with sensitive details restricted.
Triage and legal decisionPotential severity, failed barrier, GB/NI and sector applicability, responsible person, notification and timing.
Investigation recordChronology, evidence sources, facts, uncertainty, causal alternatives and worker/contractor input.
Action and hand-backInterim restrictions, stronger control selection, owner, acceptance criteria, installation and release authority.
Effectiveness and learningField test, recurrence check, risk-assessment update, cross-site communication and management decision.

Retention, access and disclosure rules depend on the record and lawful purpose. GOV.UK staff-records data protection recommends limiting access to sickness and injury details and using less intrusive absence records where practical.

How should reactive monitoring be assured?

Sample reports, investigations, decisions and field evidence to see whether people can report honestly and whether corrective controls remain effective.

Ask workers and contractors whether reports are easy to make and whether they receive feedback. Check that high-potential events reach leaders, statutory decisions are recorded, witness accounts are treated fairly, and engineering or physical controls are tested after installation. Compare repeat-event patterns with active inspections and critical-control checks.

Management should challenge overdue actions, repeated weak recommendations, changing injury definitions and apparently perfect performance that conflicts with worker experience. Close a system improvement only when reporting quality and risk control both improve, not merely when a dashboard is updated.

How should the work be managed and checked?

Use an accountable plan, practical controls, worker input and checks that show whether the controls work in real tasks.

HSE’s Plan, Do, Check, Act approach connects the risk decision to implementation, monitoring and improvement. For this topic, the duty holder should name a control owner, define the evidence that the control is effective, and change the work if that evidence fails. A management-system label or certificate is not proof that an individual exposure or task is safe. HSE managing health and safety

Official guidancePractical decisionBoundary
HSE Plan, Do, Check, ActAssign the risk owner, implement the chosen control and review field evidence after change or failure.This guidance does not replace the applicable GB or NI law, task assessment or named specialist decision.

Who needs competence for reactive monitoring?

Reporters need a trusted route; triage staff need risk and RIDDOR awareness; investigators need evidence and causal-analysis skill; managers must be able to challenge weak actions.

Practise the actual escalation for injury, near miss, ill health and critical-control failure. Investigators should know how to protect evidence, interview without blame, distinguish fact from inference, involve relevant workers and contractors, and select controls. The responsible person obtains event-specific legal advice where classification is uncertain. Train action owners to demonstrate installation and effectiveness, not merely to mark a task complete.

Choose instruction and supervised practice for the actual decision. The assessor records the demonstration and scope of authority; reassess after a change, failure or evidence that the control is not understood. A course alone does not authorise specialist work.

Role-specific learning and competence evidence
Role and learningDemonstration and assessorRecord and reassessment trigger
Workers and contractors: recognise and report injury, near miss, illness or failed control.Supervisor checks that a person can use the actual reporting and immediate-danger route.Accessible briefing and feedback; revisit after route change or missed report.
Supervisors and triage staff: protect people, preserve facts and assess potential severity.Safety lead samples a realistic first report, interim control and escalation decision.Triage record and owner; reassess after a serious near miss or classification error.
Investigators: interview fairly, reconstruct events and test causal alternatives.Competent investigation lead reviews an evidence plan, causal reasoning and control recommendation.Investigation competence and peer review; reassess after complex case or weak findings.
Responsible persons: decide GB or NI statutory reporting for the actual event.Legal or safety specialist checks a jurisdictional case and notification record.RIDDOR decision and source used; reassess after law change or missed notification.
Managers and auditors: resource stronger actions and test their effectiveness.Senior lead reviews a case through field verification and repeat-event challenge.Action test, feedback and management decision; reassess after recurring control failure.

What do the acronyms and key terms mean?

Expand a term before using it to make a safety or reporting decision; the same abbreviation can refer to different duties or units.

Acronyms and key terms
TermFull wording or expansionMeaning for this topic
RIDDORReporting of Injuries, Diseases and Dangerous Occurrences RegulationsSeparate GB and NI reporting regimes; only the responsible person reports when the applicable criteria are met.
HSE hierarchyElimination, substitution, engineering controls, administrative controls, then PPEA source-first order for choosing workplace controls; keep higher-order measures ahead of instructions or PPE.
PPEPersonal protective equipmentResidual protection; it cannot replace a missing physical barrier.
Near missEvent without actual injury but with credible harm potentialStill merits internal learning and sometimes meets a specified dangerous-occurrence test.
Dangerous occurrenceSpecified event under the applicable RIDDOR scheduleNot every near miss is reportable.
Potential severityCredible consequence under slightly different conditionsRecord separately from actual outcome.
Critical controlSafeguard needed to prevent serious harmTest its real performance, not only its existence.
Active monitoringChecking risk controls before an adverse outcomeUse alongside reactive event learning.
ERICPDEliminate, Reduce, Isolate, Control, Personal Protective Equipment, DisciplineCompare source removal and substitution with engineered separation before work rules or PPE; Discipline means checking that stronger controls keep working.