1.5 Classifying the Common Causes of Occupational Accidents

1.5 Classifying the Common Causes of Occupational Accidents

PECB requires the auditor to know the methodology used to classify the common causes of occupational accidents and the relationship of hazard and workplace. This knowledge is what allows you to judge, on the spot, whether an auditee's incident investigation under Clause 10.2 is genuine analysis or a paperwork exercise ending in "operator error".

---

1. Vocabulary First: Incident, Accident, Near Miss

ISO 45001:2018 Clause 3.35 defines an incident as an occurrence arising out of, or in the course of, work that could or does result in injury and ill health. Two notes attached to that definition carry the whole classification scheme:

• An incident where injury and ill health occurs is commonly referred to as an "accident".
• An incident where no injury and ill health occurs, but has the potential to do so, may be referred to as a "near miss", "near hit" or "close call".

The examinable consequence: ISO 45001 has one operative term, "incident", and it covers both accidents and near misses. An organization whose procedure requires investigation only where a person was hurt has narrowed the scope of Clause 10.2 and is nonconforming — near misses are incidents and must be within the process.

Hazard and workplace

Clause 3.19 defines a hazard as a source with a potential to cause injury and ill health. Clause 3.6 defines a workplace as a place under the control of the organization where a person needs to be or go for work purposes. The relationship examined by PECB is one of exposure: a hazard produces risk only where a worker's presence in a workplace creates exposure to it. This is why hazard identification under Clause 6.1.2.1 must consider workplace design, equipment, work organization, social factors, and situations arising outside the immediate workplace that can adversely affect health and safety within it.

---

2. The Cause Hierarchy: Immediate, Underlying, Root

The dominant classification methodology sorts causes into three tiers. Auditors use it as a scoring rubric against any investigation report.

Tier 1 — Immediate (direct) causes

The acts and conditions present at the moment of the event.

• Unsafe acts (behaviour): operating without authority, bypassing an interlock, failing to isolate energy, removing a guard, working at speed, incorrect PPE use, horseplay.
• Unsafe conditions (physical state): defective tools, inadequate machine guarding, poor housekeeping, insufficient ventilation, inadequate illumination, excessive noise, exposed live conductors, unstable stacking.

Tier 2 — Underlying (basic) causes

Why the act or condition existed.

• Personal factors: inadequate physical or mental capability, lack of knowledge or skill, stress, fatigue, improper motivation, unfamiliarity with the task.
• Job/system factors: inadequate supervision, deficient engineering design, defective procurement, inadequate maintenance, worn tools, poor work standards, abuse or misuse tolerated over time.

Tier 3 — Root causes

The management system deficiencies that allowed the underlying causes to persist: absent or inadequate standards, standards not communicated, standards not enforced, no verification of compliance, no competence assessment, no management of change, no worker consultation mechanism.

Only Tier 3 generates a corrective action that prevents recurrence. Correcting Tier 1 is correction, not corrective action — a distinction ISO 45001 Clause 10.2 draws explicitly and one the exam tests repeatedly.

---

3. The Three Causation Models You Are Expected to Recognise

Heinrich's domino theory and the 300:29:1 ratio (1931)

H. W. Heinrich modelled an accident as five falling dominoes: ancestry/social environment → fault of person → unsafe act or unsafe condition → accident → injury. Removing the middle domino (the unsafe act/condition) stops the sequence. From reviewing insurance claims he proposed that for every 1 major injury there are 29 minor injuries and 300 no-injury accidents.

• Its contribution: it created the case for investigating near misses — the 300 events are free information about the mechanism that will eventually produce the 1.
• Its defect: Heinrich attributed roughly 88% of accidents to unsafe acts by people, which drives blame-based investigation. Modern practice, and ISO 45001, reject that emphasis.

Bird's accident ratio (1969)

Frank E. Bird Jr. analysed 1.75 million incident reports across 297 companies and produced the ratio 1 serious injury : 10 minor injuries : 30 property damage accidents : 600 near misses. Bird's contribution was the loss causation model, adding lack of control by management as the first domino — the direct conceptual ancestor of a management-system standard.

Reason's Swiss cheese model (1990)

James Reason distinguishes:

• Active failures — unsafe acts by people in direct contact with the system, whose effects are felt immediately.
• Latent conditions — "resident pathogens" created by decisions taken far upstream: understaffing, poor design, unworkable procedures, production pressure, deficient training budgets.

Defences are layers of cheese; holes are weaknesses that open and close. An accident occurs when holes in successive layers momentarily align. Latent conditions are the root causes; active failures are the immediate causes. This is the model that best explains the ISO 45001 requirement in Clause 10.2 to determine whether similar incidents exist or could potentially occur, and to review the effectiveness of existing controls rather than the conduct of the injured person.

---

4. Audit Application

When sampling incident records, test the investigation against the hierarchy:

1. Scope test: does the process capture near misses, ill health, and incidents involving contractors and visitors — or only lost-time injuries to employees?
2. Depth test: does the analysis reach a management system cause, or terminate at the worker?
3. Consultation test: Clause 5.4 requires participation of non-managerial workers in incident investigation. Was the workforce involved, or was the report written by the safety department alone?
4. Hierarchy test: Clause 10.2 requires the organization to determine and implement action, including corrective action, in accordance with the hierarchy of controls. An action plan consisting only of retraining and a toolbox talk sits entirely at the administrative tier and has not applied the hierarchy.
5. Effectiveness test: was the effectiveness of the corrective action reviewed, and were the risks and opportunities re-evaluated?

Worked example. A worker suffers a hand laceration clearing a jam in a conveyor. The investigation records the cause as "employee reached into machine while running — disciplinary warning issued". An auditor applies the hierarchy: the immediate cause is the unsafe act; the underlying cause is that jams occur several times per shift and the isolation procedure takes eleven minutes, creating production pressure; the root cause is that no management of change was performed when throughput was increased and the guard interlock design was never reassessed. The corrective action offered addresses only the immediate cause and sits at the administrative tier. The auditor raises a nonconformity against Clause 10.2 — the organization did not determine the root causes of the incident and did not act in accordance with the hierarchy of controls — and notes the Clause 8.1.3 management-of-change failure as a linked audit trail.