8 questions from Episode 11 with Dr. Thomas Krause

Episode 11 of Headline Podcast, published on December 3, 2025, brought Dr. Thomas Krause into conversation with Andreza Araujo and Dr. Megan Tranter about leadership, trust, culture, and incident analysis. The central thesis he defended is that safety leaders learn more when they investigate the conditions that shaped behavior than when they stop at the employee action visible at the end.

1. What made the procedure hard to follow?

An incident review should ask what made the procedure hard to follow before it asks why the worker failed to follow it. Episode 11 matters because Dr. Krause does not treat procedure deviation as a simple moral defect. He points leaders toward the work system that made the written expectation difficult, slow, unclear, unavailable, or inconsistent with the actual task.

On Headline Podcast, Dr. Thomas Krause said: 'Incident analysis always looks like it was the employee's fault for not following the procedure, but look deeper and you see following it was made very difficult by system factors set by decisions made a year, or five years, ago.' That sentence should change how EHS managers read a file. The question is not whether the rule existed. The question is whether the organization made the rule workable under pressure.

OSHA describes hazard identification and assessment as a process that includes worker input, inspections, incident investigations, and review of hazards associated with routine and non-routine work. That view supports a practical investigation standard: compare the document with the job as performed, including time, tools, staffing, access, supervision, and competing instructions.

This question also connects with Headline's guide on building an incident evidence map in 48 hours. If the evidence map contains only the final action, it is too thin for a serious investigation.

2. Which leadership decision shaped the event long before the event?

Many incident causes sit upstream of the day of injury. A staffing decision, capital deferral, maintenance backlog, production target, contractor selection, training shortcut, or supervisor span of control can create the conditions that later appear as individual error. Episode 11 is useful because it treats leadership as part of the causal evidence, not as background context.

On Headline Podcast, Dr. Thomas Krause said: 'We were surprised to learn the strongest predictor of success was the quality of leadership given to the initiative.' He was discussing safety initiatives, but the same logic applies after an incident. Leadership quality affects what work gets funded, what weak signals get escalated, and what corrective actions survive the first month after the review closes.

James Reason's work on latent failures remains a strong anchor for this question because serious events often emerge from decisions made far from the point of contact. The investigator should therefore ask which decisions created the exposure path, which decisions kept it open, and which decisions made the final worker action more likely.

A useful review sample is 10 recent investigations. For each one, identify at least 2 management decisions that shaped the event. If the team cannot name any, the method may be collecting facts while avoiding power.

3. Did the investigation preserve trust or teach silence?

An investigation either preserves trust or teaches the workforce to protect itself from the next investigation. Episode 11 is especially relevant for EHS leaders because Dr. Krause links trust with the quality of safety information. People do not give leaders the full story when they believe the story will be used mainly to identify a culprit.

Trust can be damaged by small signals. A supervisor interrupts an interview. A manager asks who messed up before asking what failed. A corrective action names retraining before the evidence has been reviewed. Each move teaches workers that honesty is risky, even when the company says it wants learning.

OSHA identifies worker participation as a core part of safety and health programs, including ways for workers to report hazards and participate without fear of retaliation. Incident investigation should meet that same standard. Workers should be able to describe pressure, ambiguity, and flawed controls without being treated as disloyal.

Andreza Araujo's work in The Illusion of Compliance makes the same point through a culture lens. A clean report can still hide an unsafe culture when workers learn to provide the version of events leaders prefer.

4. What did the trust score hide?

Average trust scores can hide risk because the people who distrust leadership may be exactly the people closest to the exposure. Episode 11 gives leaders a useful warning about comfort with averages. A high percentile position does not mean every crew, shift, or supervisor relationship is safe enough for truth.

On Headline Podcast, Dr. Thomas Krause said: 'You could have 60% of people say they trust their supervisor and still be in the 90th percentile, so the leader pats himself on the back while 40% of his people don't trust their boss.' In an incident investigation, that 40 percent matters. They may hold the missing evidence, the earlier warning, or the reason a concern never reached management.

EU-OSHA explains that leadership and worker participation reinforce each other when management creates real dialogue about occupational safety and health. The investigation should therefore segment trust evidence by crew, shift, role, contractor status, and supervisor rather than relying on a broad culture score.

The practical test is simple. In the next serious review, compare formal statements with informal field interviews across 2 shifts and at least 3 roles. If the official narrative is cleaner than the field narrative, the investigation may be hearing status more than truth.

5. What would the status quo investigation miss?

A status quo investigation usually finds the visible deviation, the breached rule, the missing signature, or the person closest to the harm. A stronger Episode 11 lens asks what that explanation misses. The difference is not academic, because a shallow explanation produces shallow corrective action.

The table also explains why operator blame in RCA remains such a persistent problem. Blame gives the organization a fast story. Investigation gives the organization a harder decision.

6. Which corrective action changes work instead of language?

A corrective action should change work, not only the wording around work. Episode 11 pushes leaders to ask whether the action removes the condition that made the event possible. If the answer is only retraining, coaching, communication, or a procedure update, the team should test whether knowledge was truly the missing control.

Training is sometimes necessary, but it is often too convenient. A maintenance access problem does not disappear because the crew attends a briefing. A production pressure pattern does not change because the investigation recommends renewed awareness. A confusing permit interface does not become clear because the worker signs a lesson-learned document.

Andreza Araujo's Far Beyond Zero challenges leaders to look beyond clean numbers and administrative closure. That argument fits incident review because the important result is not that the investigation closed in 30 days. The important result is that the exposure path became harder to repeat.

Use 3 proof questions before closure: what condition changed, where was it verified, and what repeat signal will show whether the fix failed. If the action cannot answer those questions, it may be complete as a task but weak as risk control.

7. Did the review protect the original worker language?

Incident reports often become safer for leaders as they move upward. Worker language is specific, uncomfortable, and operational. Executive language can become polished, abstract, and less useful. Episode 11 should push EHS managers to protect the original risk language long enough for decision makers to hear it.

A worker may say the job was rushed, the tool was wrong, the supervisor was unavailable, or the previous warning was ignored. By the time that statement reaches the monthly review, it may appear as a communication opportunity, a behavior concern, or a need to reinforce standards. The meaning has changed.

This translation loss is visible in many investigations. It is also preventable. Keep a short evidence appendix with exact worker phrases, time stamps, job photos, planning records, permit notes, and supervisor decisions. Do not publish confidential or identifying details, but do preserve enough operational language for leaders to feel the real problem.

Headline's article on incident chain of custody expands this point. Evidence is not only physical. Language can also degrade when too many people smooth it before leadership sees it.

8. Recommendation

EHS leaders should use Episode 11 to run a 30-day investigation-quality review. Choose 5 recent investigations, including at least 1 serious-potential near miss, and test whether each file explains the work system that made the final action likely. The review should include field evidence, worker language, leadership decisions, and proof that corrective actions changed conditions.

Start with 8 questions from this article. What made the procedure hard to follow? Which leadership decision shaped the event early? Did the process preserve trust? What did the trust score hide? What would the status quo miss? Which action changed work? Did the review protect worker language? What will leaders verify after closure?

The review should not become another audit checklist. Its value is in the conversation it forces with operations, maintenance, HR, legal, contractors, and senior leadership. If the file points mostly to the worker, but the field points to planning, pressure, tools, supervision, or decision rights, the investigation is not finished.

Episode 11 with Dr. Thomas Krause gives leaders a practical standard for incident review: do not stop at the behavior that made the event visible. Listen to the full conversation: Listen to the full conversation.