Causal Factor Analysis: 5 Myths Boards Still Believe

OSHA's incident investigation guidance warns that incidents usually involve equipment, procedural, training, and safety-program deficiencies, not a single visible mistake. This article challenges five myths about causal factor analysis so boards and senior EHS leaders can demand investigations that change risk, not merely close reports.

On the Headline Podcast, Andreza Araujo and Dr. Megan Tranter frame safety as a real leadership conversation, which matters after a serious event because the first executive question can either widen or narrow the investigation. Causal factor analysis is useful only when it protects evidence, separates conditions from opinions, and forces leaders to ask which controls were weak before the incident made that weakness public.

Why causal factor analysis fails in serious incident reviews

Causal factor analysis fails when leaders treat it as a reporting technique instead of a governance discipline. The method should map the events and conditions that contributed to the incident, although many organizations compress it into a neat cause statement because the board wants closure before the operational picture is mature.

The Federal Transit Administration describes causal factors as part of structured safety investigations, and OSHA's public guidance points leaders toward underlying program deficiencies rather than immediate blame. Those references matter because they protect the investigation from a narrow story in which the last person near the job becomes the whole explanation.

As co-host Andreza Araujo explores in Safety Culture: From Theory to Practice, culture shows up in what an organization permits, measures, corrects, and rewards under pressure. After a serious event, that means the investigation must test the pressure system around the work, not only the worker's last decision.

1. Myth: the causal factor is the root cause

A causal factor is an event or condition that contributed to the incident, while a root cause claim should explain why the condition existed and why controls failed to correct it. Treating those two ideas as identical creates shallow findings because the investigation stops at the first useful label.

The myth feels efficient because executives want a single sentence they can take to the board. Yet serious incidents rarely come from one cause. A dropped load, a chemical exposure, or a confined-space event usually combines planning quality, field verification, supervision, equipment condition, contractor interface, and production pressure.

The practical test is whether the finding points to a control decision. If the report says poor communication, ask which communication channel failed, whose decision rights were unclear, what information was unavailable, and where similar work still depends on the same weak channel.

This is where operator blame in RCA becomes dangerous. A worker choice may be visible, although it is not necessarily the causal factor that leaders can control across the system.

2. Myth: a good timeline proves causality

A timeline proves sequence, not causality, unless it also shows the conditions that made each step possible. Many incident timelines look precise because they contain times, names, and events, but precision is not the same as causal logic.

The common trap is chronological confidence. A report may say the permit was signed at 08:05, the job began at 08:18, the alarm sounded at 08:42, and the injury occurred at 08:44. That sequence matters, although it does not explain whether the permit checked the right exposure, whether the alarm was trusted, or whether supervision had authority to stop the task.

James Reason's work on organizational accidents is useful here because it separates active failures from latent conditions. A causal timeline must therefore include the earlier decisions that shaped the final work environment, including maintenance backlog, staffing, design compromise, weak handover, unclear escalation, and tolerated bypasses.

Senior EHS leaders should require one extra layer in every serious incident timeline: mark each moment where a control was selected, verified, degraded, challenged, bypassed, or ignored. That turns a sequence into an investigation map.

3. Myth: witness statements are enough evidence

Witness statements are essential, but they are not enough evidence because memory changes under stress and people interpret events through their role, fear, and available information. A serious incident investigation needs statements, scene facts, documents, electronic records, photographs, task planning records, and control verification evidence.

The myth persists because witness interviews are easier to schedule than hard evidence is to preserve. A supervisor can interview three people in one afternoon, while preserving a damaged component, extracting equipment data, freezing a permit trail, or reconstructing a control state may require technical authority and operational disruption.

On Headline Podcast, the recurring leadership question is whether safety professionals are prepared to have harder conversations before the story becomes convenient. In causal factor analysis, that means asking whether the evidence supports the narrative or whether the narrative arrived first.

Use witness interviews to understand perception, sequence, uncertainty, and competing accounts. Then test those accounts against physical and system evidence before naming a causal factor.

4. Myth: every causal factor needs one corrective action

One causal factor may require several actions because the contributing condition can sit across engineering, maintenance, supervision, procurement, training, planning, and governance. The common one-to-one action logic looks tidy, although it often leaves the real barrier weakness untouched.

For example, if the causal factor is ineffective isolation verification, retraining the crew may only address one layer. The organization may also need a revised isolation standard, field test criteria, supervisor sign-off, contractor qualification, maintenance of isolation points, and a restart rule for deviations.

5 separate ownership questions should be answered before closing an action: who owns the barrier, who verifies it, who funds the fix, who can stop the work, and who confirms effectiveness after restart. If those answers all point to EHS, the action plan is probably misassigned.

This connects directly to barrier failure review. Causal factor analysis names contributing conditions, while barrier review tests what should have stopped the exposure and who owns that control in real work.

5. Myth: closing the report closes the risk

Closing the report does not close the risk unless leaders verify that the failed condition cannot credibly repeat in similar work. A final report may satisfy governance rhythm, but the risk remains open when actions are administrative, untested, or limited to the exact incident location.

The board-level problem is false completion. A dashboard may show the investigation as closed, actions as assigned, and due dates as green, while similar assets, shifts, contractors, or tasks still carry the same exposure. This is how an organization can look responsive and still remain fragile.

30 days after restart, senior leaders should ask for verification evidence from similar operations, not only from the incident site. That evidence may include field observations, control tests, maintenance records, permit quality checks, and supervisor challenge logs.

The better closure question is not whether the report is finished. It is whether the organization has found where else the same causal pattern can live, especially when the original event had serious incident potential.

What boards should demand instead

Boards should demand causal factor analysis that separates evidence, conditions, control failure, ownership, and recurrence risk. The investigation does not need theatrical certainty, but it does need enough discipline to show what leaders will change before the same exposure appears elsewhere.

Dr. Megan Tranter's work on clarity in chaotic leadership moments fits this point because post-incident governance is full of urgency, legal sensitivity, operational pressure, and emotional noise. Clarity means the report says what is known, what is inferred, what remains uncertain, and which control decision cannot wait.

In a recent Headline Podcast conversation about visible felt leadership, the leadership message was not about performing concern after harm. It was about making better decisions while the work is still messy, which is exactly the standard causal factor analysis should meet after a serious event.

Each week that a serious incident report sits in draft without testing recurrence across similar work leaves leaders with an explanation of the past, while the same exposure may still be active on another shift, line, site, or contractor interface.

Causal factor analysis: myth vs disciplined practice

Conclusion

Causal factor analysis becomes valuable when it stops chasing a neat cause and starts testing the conditions that made the incident possible. The board's role is to protect that depth, especially when operational pressure rewards quick closure.

Use the next serious incident review to ask five questions: what conditions contributed, which controls failed, what evidence supports the claim, who owns the barrier, and where else the same pattern could repeat. Without those answers, the organization has a report, but it does not yet have risk reduction.