How Dr. Thomas Krause Thinks About the Safety Behaviors Leaders Create

Episode 11 of the Headline Podcast, published on December 3, 2025, featured Dr. Thomas Krause discussing why leadership quality changes safety performance.

His central argument was that leaders create the conditions in which people either repeat safe behavior or learn that another behavior is more rewarding.

Safe behavior begins with the conditions leaders create

Safe behavior is more reliable when leaders make the desired action practical under real operating pressure. A worker may know the rule, agree with the rule, and still bypass it when the system withholds time, information, equipment, authority, or support. The useful leadership question is not only whether people complied. It is whether the operating conditions made the safer choice the easier choice to sustain.

That distinction moves the conversation away from personality. A worker who rushes a check may be displaying poor judgment, but the event may also show a schedule that leaves 2 minutes for a task that needs 10, a handoff that omits 3 critical facts, or a supervisor who treats delay as failure.

Krause's point is especially relevant to safety professionals who are asked to improve behavior through campaigns. A poster can describe the desired action, yet only a leader can protect the time and decision rights that let the action survive a difficult shift.

The OSHA management leadership guidance makes a similar connection by describing leadership through commitment, resources, participation, and accountability. The practical test is whether those elements appear in the next decision, not whether the organization has published them.

The strongest predictor is the behavior leaders reward

Leadership quality predicts safe behavior because workers watch consequences more closely than slogans. When a leader protects a stop, funds a correction, and asks for bad news without humiliation, the team receives evidence about what the organization truly values. When the same leader praises speed after a shortcut, the shortcut becomes the more credible instruction.

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."

The quote is not a request for more visible executive presence. It is a request to inspect the small decisions that train a workforce. Which behavior receives immediate recognition? Which concern waits 3 weeks for an answer? Which supervisor is quietly protected after a production miss, and which one is blamed after a safety delay?

Andreza Araujo's Headline Podcast work is valuable here because it keeps leadership connected to operating detail. Across more than 25 years of multinational EHS leadership and 250-plus cultural transformation projects, her editorial position is strongest when it asks how a stated value becomes a decision that workers can feel.

A useful review samples 5 recent decisions rather than 5 campaign materials. Compare the response to a stop-work request, a staffing shortage, an overdue repair, a changed condition, and a near miss. The pattern will usually reveal which behavior the organization is actually teaching.

Trust changes what workers are willing to do next

Trust affects safe behavior before a task begins because it determines whether workers will disclose uncertainty, ask for help, and challenge a weak plan. A worker who expects embarrassment or delay may continue silently, even when the formal process says that speaking up is welcome.

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."

The value of the 60% example is not the ranking. It is the blind spot. An average can make a leader feel successful while a large minority learns to withhold information. That minority may include the newest worker, the contractor with the least job security, or the experienced operator who has already seen concerns disappear without explanation.

Leaders should therefore ask 4 separate questions. Can people raise a concern? Do they receive acknowledgment? Does the decision owner explain what happened? Does the person who raised the issue remain protected after the decision? A positive answer to the first question does not prove that the other 3 exist.

The OSHA worker participation guidance supports this distinction by connecting participation with access to information, the ability to report hazards, and protection from retaliation. Safe behavior becomes more likely when participation changes the work rather than only collecting comments.

Incident reviews must examine the behavior made difficult

An incident review should ask why the expected behavior was difficult to perform before it asks why the worker failed to perform it. That sequence does not remove personal accountability. It identifies the conditions that made the behavior unstable, which is necessary if the organization wants the same event to become less likely.

Krause described a familiar pattern in which an investigation ends with the employee's failure to follow a procedure. The investigation may be technically complete while missing the decisions that shaped the procedure, its usability, and the pressure surrounding the 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 system factors set by decisions made a year, or five years, ago."

The review should reconstruct at least 6 conditions: the information available, the time allowed, the tools provided, the supervisor's instructions, the competing production demand, and the escalation route. Each condition can turn a safe behavior into a fragile exception.

The UK HSE investigation guidance recommends looking beyond the immediate event toward underlying causes and management arrangements. That lens helps a safety team distinguish a genuine knowledge gap from a work-design gap that training cannot repair.

What the guest's view changes in a behavior program

Krause's view changes a behavior program by moving its first intervention from observation to operating design. Before asking workers to repeat a behavior, leaders should identify the trigger, the available time, the authority to pause, and the response that follows a concern.

A behavior observation can still be useful, but the observation should record the environment around the action. If a worker skips a verification, record whether the task started late, whether the permit was clear, whether the control owner was reachable, and whether the supervisor had already signaled that the job could not slip.

This also explains why a training completion rate can look strong while field behavior remains unstable. Training may reach 95% of the workforce, yet the job may provide no protected time to use the skill. The metric describes exposure to information, not the credibility of the operating condition.

For a practical next step, compare this approach with behavioral feedback after a control deviation, then use the questions that expose false confidence before treating a quiet task as a reliable one.

Before and after the leadership question

The leadership question becomes stronger when it changes from "Why did the worker do that?" to "What did the work system teach the worker to do next?" The second question still allows a precise review of choices, yet it also reveals the incentives and constraints that make those choices repeat.

The table does not replace an investigation or a behavior conversation. It changes the quality of the questions that come first, which is where many reviews either open the system or close it around a convenient individual explanation. That change also gives supervisors a clearer basis for coaching the next decision.

In Andreza Araujo's safety-culture vocabulary, this is the difference between declaring a value and making the value operational. A leader who wants safe behavior must make the safer choice visible, possible, and worth repeating.

Recommendation

Use Episode 11 as a one-week leadership test. Select 3 recurring safety behaviors, observe the conditions around each one, interview 4 workers about the response they expect, and review 2 recent incidents for system factors that made the expected behavior difficult.

Do not begin by launching another campaign. Begin by naming the decision that workers must make, the support that decision requires, and the consequence leaders will protect when the safer choice takes longer.

Then compare the findings with the task-familiarity review so that routine work is tested for changed conditions rather than treated as self-explanatory. If the same barrier appears in all 3 behaviors, assign the correction to the role that controls the barrier instead of assigning the gap to the worker who encountered it.

Listen to the full conversation with Dr. Thomas Krause on the Headline Podcast, then ask which behavior your next leadership decision will create.

Safety culture becomes credible when people can predict what leadership will protect under pressure. Continue the conversation at Headline Podcast.