How to Run a Fit-for-Work Review Before High-Risk Tasks

A fit-for-work review is not a medical diagnosis at the gate. It is a practical control before high-risk work, because the person who looks present on the roster may still be too fatigued, distressed, medicated, distracted, or overloaded to perform a task where one wrong movement can defeat several barriers.

The procedure below is built for supervisors, EHS managers, and operations leaders who need a repeatable review before energized work, confined-space entry, line breaking, lifting, driving, mobile-equipment operation, work at height, chemical transfer, or any task where a moment of poor attention can become a serious injury or fatality. OSHA's Job Hazard Analysis guide, OSHA 3071, frames safe work around task steps, hazards, and controls. ISO 45001:2018 also expects organizations to identify hazards and address OH&S risks through operational controls. A fit-for-work review connects those management-system requirements with the human condition of the crew before the job begins.

What you need before starting

The supervisor needs three inputs before the conversation starts: the task risk level, the crew's recent work pattern, and a clear escalation route. Without those inputs, the review becomes a personal interview with no operational authority, which is why many companies either avoid the conversation or mishandle it.

Across 25+ years leading EHS in multinational environments, Andreza Araujo has observed that weak pre-task reviews usually fail before the first question is asked. The company asks supervisors to notice fatigue or distress, although the same supervisors cannot change the work plan, delay the task, call occupational health, or ask for another competent person. As Andreza Araujo argues in Safety Culture: From Theory to Practice, culture shows up in repeated operational decisions, not in the sentence printed on a poster.

Step 1: Define which tasks require the review

Do not run the same review for every routine activity. Start with tasks where impaired attention, slower reaction, or poor judgment can break a critical control. The practical list usually includes energy isolation, confined-space entry, mobile equipment, driving, line breaking, hot work, lifting, work at height, emergency response, hazardous chemical handling, and work performed alone.

The review should also trigger when work conditions change, because the same person may be fit for routine inspection but not fit for a late-night intervention after overtime. This distinction matters for fairness. The question is not whether the worker is a good employee. The question is whether today's condition matches today's exposure.

Step 2: Check the work pattern before the worker

Before asking personal questions, look at the work system. Review the last shift, overtime, consecutive nights, quick returns, call-outs, commute exposure, and whether the person is entering a task after a demanding period. NIOSH describes workplace fatigue as commonly associated with nonstandard schedules that shorten or disrupt sleep, and OSHA's worker-fatigue guidance also connects extended or irregular shifts with higher injury and accident risk.

This step prevents a common injustice. If the roster has created the exposure, the answer cannot be to tell the worker to be more careful. The review should connect with workload risk indicators and with the existing fatigue review for sleep disorders in shift workers, because the same work-design pattern often appears in several dashboards.

Step 3: Ask observable questions, not clinical questions

Supervisors should not diagnose anxiety, depression, sleep disorders, medication effects, or substance use. They should ask questions linked to safe task execution: whether the person feels alert enough for the work, whether anything may affect concentration, whether medication or poor sleep could interfere with the task, and whether the person needs occupational health support before starting.

The wording must protect dignity. A supervisor who asks, "Are you mentally okay?" may open a conversation that they cannot handle. A better question is, "Is there anything today that could affect your attention, reaction, balance, or decision-making during this task?" That wording keeps the discussion anchored in operational control.

Step 4: Separate privacy from safety-critical disclosure

The worker does not need to disclose a diagnosis for the supervisor to manage the exposure. If a medication causes drowsiness, if a family crisis has kept the worker awake, or if a panic episode has left the person unable to concentrate, the safety-relevant fact is the functional effect on the task, not the private medical detail behind it.

This boundary protects both sides. The supervisor avoids collecting sensitive information that belongs with occupational health or HR, while the company still acts on the operational risk. In more than 250 cultural transformation projects supported by Andreza Araujo's team, this distinction often separates respectful control from intrusive management.

Step 5: Compare the person's condition with task criticality

A fit-for-work review is not a yes-or-no test of the person. It is a match between current condition and task exposure. A worker with mild tiredness may be able to do low-risk housekeeping under normal supervision, although the same worker should not operate mobile equipment, enter a confined space, or perform electrical switching without additional controls.

This is where many organizations lose the plot. They ask whether the worker is fit in general, then ignore the specific task. James Reason's work on latent failures helps explain the risk, because the visible error can happen at the front line while the deeper weakness sits in planning, staffing, and task assignment.

Step 6: Decide the control before the conversation ends

The review must produce a decision that the supervisor can execute immediately. The decision may be to proceed, add a second checker, reassign the person, delay the task, move the task to daylight, involve occupational health, reduce duration, add rest time, or activate stop-work authority.

If the only answer is "be careful," the review has failed. The company should connect this step with stop-work authority, because a worker who reports poor readiness should not be punished for making a risk visible before harm occurs.

Step 7: Record the exposure without writing a medical file

The record should document task, date, reviewer, risk trigger, operational decision, and control used. It should not include diagnosis, private medical details, family details, or speculation about personal causes. If occupational health becomes involved, the medical record belongs in the proper confidential channel.

This discipline protects data quality. It also helps the EHS manager identify patterns such as repeated overtime before critical work, frequent reassignment after night shift, specific crews with high fatigue signals, or supervisors who proceed too often despite visible concerns.

Step 8: Build escalation rules for red flags

Some findings require immediate escalation. Examples include visible impairment, confusion, loss of balance, inability to stay awake, severe emotional distress, suspected substance influence, medication effects that impair alertness, or a worker stating that they cannot safely perform the task.

The escalation rule should name who must be called and what happens to the task while the call is made. A review that identifies a red flag but leaves the job moving creates false confidence. The EHS manager should treat this the same way they treat a failed permit condition or a missing isolation verification.

Step 9: Verify the control during the job

Fitness can change during work. Heat, time pressure, long duration, abnormal events, alarms, conflict, and repeated decisions can reduce the margin after the task has started. The supervisor should plan a mid-task check when the work lasts longer than expected or when the most critical step happens late in the sequence.

This is especially important where decision fatigue is already visible. The review before work is only the first barrier. The field verification confirms that the control still exists when the task reaches its most dangerous point.

Step 10: Review trends monthly with EHS, HR, and operations

A monthly review should look for repeated patterns, not individual blame. Useful questions include which tasks trigger most concerns, which shifts produce more fatigue signals, whether overtime appears before high-risk work, whether supervisors use the escalation route, and whether reassignment creates staffing pressure elsewhere.

Andreza Araujo's safety culture work emphasizes that leadership decisions shape behavior before the worker reaches the field. If the monthly review shows that the same crew is always tired before high-risk work, the root issue may be staffing, planning, shutdown compression, transport, or production pressure rather than personal discipline.

Decision table for supervisors

Common mistakes that make the review unsafe

The first mistake is treating the review as a worker confession instead of a task-control decision. The second is collecting private health information that the supervisor does not need. The third is recording a concern but sending the same person into the same high-risk task without changing the control set.

The fourth mistake is separating fit-for-work from critical control verification. A tired worker is not the only risk. The real exposure appears when fatigue, weak supervision, poor planning, and a fragile control meet in the same job.

What good looks like after 30 days

After 30 days, the company should have a short list of high-risk tasks, a supervisor script, an escalation route, a privacy-safe record, and a monthly trend review. The review should be short enough to run before work, but serious enough to change the plan when the person's condition and the task exposure do not match.

Fit-for-work only becomes credible when workers see that speaking up changes the job instead of harming their reputation. That is the cultural test. A company that protects the person before the task starts is not lowering standards. It is proving that production pressure does not outrank life-critical control.