ISO 45001 Clause 5.4: Worker Consultation & Participation

Clause 5.4 is the clause where an ISO 45001 system either belongs to the people doing the work or belongs to the safety manager. In my experience, many of the audit findings against it have nothing to do with a missing procedure. The procedure exists. It is just that nobody on the floor has ever used it, and the records show managers talking to managers.

This guide walks through what clause 5.4 of ISO 45001:2018 actually says, how consultation differs from participation, which topics the standard singles out, and what evidence an auditor will ask for. If you want the shorter overview first, our post on worker participation in ISO 45001 and what clause 5.4 requires covers the basics. This is the longer reference.

What does ISO 45001 clause 5.4 require?

The lead sentence of clause 5.4 requires the organization to "establish, implement and maintain a process for consultation and participation of workers at all applicable levels and functions, and, where they exist, workers' representatives, in the development, planning, implementation, performance evaluation and actions for improvement of the OH&S management system."

Two things in that sentence deserve attention before you get to the lettered list.

First, it asks for a process, not a committee and not a suggestion box. A committee can be part of the process, but the standard wants something you can describe, run, and show evidence from.

Second, the scope runs across the whole management system lifecycle: development, planning, implementation, performance evaluation, and improvement. Plenty of organizations consult workers during hazard identification and then never involve them again until an incident. Clause 5.4 does not allow that narrow reading.

The clause then breaks into five lettered requirements (a to e) plus four notes. I will take them in order.

Who counts as a "worker" under clause 5.4?

The definition in clause 3.3 is broad: a person performing work or work-related activities under the control of the organization. That includes top management, managerial and non-managerial persons, and it reaches people who are not employees, such as agency staff, contractors working under your control, and volunteers.

This matters because the clause says "workers at all applicable levels and functions." If your system only engages direct employees, you have left out part of the population the standard covers. I would start any 5.4 gap assessment by listing every category of person who works under your control and asking how each one is currently consulted.

What is the difference between consultation and participation?

The standard defines these separately in clauses 3.4 and 3.5, and the difference is practical. Participation is involvement in decision-making. Consultation is seeking views before making a decision. In plain terms, when you consult workers, you ask and then management decides. When workers participate, they help make the decision.

The split mirrors how the standard itself divides its two lists of topics. Items in 5.4 d (policy, objectives, legal compliance, procurement, audit programme and similar) are about consultation. Items in 5.4 e (hazard identification, controls, training, incident investigation and similar) are about participation. That is not an accident, and in my view it is a useful design: management keeps final say over strategic items such as policy, while the people closest to the hazards get a real seat on the operational ones.

What does clause 5.4 a) to c) require you to provide?

The first three lettered items are the enabling conditions. Without them, the lists in d) and e) are mostly paper.

5.4 a) Mechanisms, time, training and resources

You must provide the mechanisms, time, training and resources necessary for consultation and participation. Note 1 adds that worker representation can be one such mechanism.

The word to underline is time. If your toolbox talk is scheduled for the last five minutes of a shift when everyone is putting tools away, you have a mechanism but not time. If a worker joins an incident investigation team and their production targets stay the same, you have given them a role without resources. Auditors who understand the clause will ask about this.

5.4 b) Timely access to clear, understandable and relevant information

Workers need timely access to information about the OH&S management system that is clear, understandable, and relevant. This is the bridge to clause 7.4 on communication. A 40-page procedure written in dense compliance language does not meet it. A one-page summary of the top five risks in a task, in the languages your workforce actually speaks, probably does.

5.4 c) Removing obstacles and barriers

You must determine and remove obstacles or barriers to participation, and minimize those you cannot remove. Note 2 names examples directly: failure to respond to worker inputs or suggestions, language or literacy barriers, reprisals or threats of reprisals, and policies or practices that discourage or penalize worker participation.

The first example on that list is the one I see most often. A worker raises a concern, hears nothing back, and learns that raising concerns is pointless. Closing the loop on every input, even when the answer is "we considered it and decided not to," is one of the cheapest and most effective things you can do.

Reprisal is the heavy one. For US sites only, section 11(c) of the Occupational Safety and Health Act prohibits discrimination against employees for exercising safety and health rights, and 29 CFR 1904.35(b)(1)(iv) separately prohibits discriminating against an employee for reporting a work-related injury or illness. If your disciplinary practices could be read as penalizing people for reporting, that is both a compliance problem and a clause 5.4 c) barrier. These OSHA provisions do not apply outside the United States, where local law governs.

Which topics must you consult non-managerial workers on? (5.4 d)

Clause 5.4 d) says the organization shall emphasize consultation of non-managerial workers on nine topics. Here they are, with what I would expect to see as evidence for each. The topic wording is paraphrased for brevity, so check the exact text of clause 5.4 d) in the standard before citing it in an audit.

A quick note on item 9, since "continual improvement" is a frequent search. In ISO 45001 it means a recurring process of enhancing OH&S performance, and the standard (clause 3.37) ties it to achieving improvement in OH&S performance consistent with the OH&S policy. Clause 10.3 requires you to improve continually the suitability, adequacy and effectiveness of the system, and 10.3 c) specifically calls for participation of workers in implementing actions for improvement. That is where clause 5.4 and clause 10 meet.

Item 6 is the one organizations most often skip. Procurement of equipment and decisions about contractors affect the people who will use the equipment or work beside the contractor, yet those decisions usually sit in purchasing or operations without any frontline voice.

Which topics must non-managerial workers participate in? (5.4 e)

Clause 5.4 e) lists six topics where the standard wants real participation, meaning involvement in the decision itself.

1. Determining the mechanisms for their own consultation and participation. Workers help design the process described in the lead sentence. This is the item people forget, and it is a good one. If you built your participation process in a management meeting, a gap exists here.
2. Identifying hazards and assessing risks and opportunities. This ties directly to clauses 6.1.2.1 and 6.1.2.2. Clause 6.1.2.1 requires a process for ongoing and proactive hazard identification, and items a) to g) of that sub-clause list what the process must take into account, including how work is organized and social factors (a) and routine and non-routine activities and situations (b). Frontline workers are usually the best source on what actually happens versus what the procedure says. For methods, see our guide to job hazard analysis as an ISO 45001 tool.
3. Determining actions to eliminate hazards and reduce OH&S risks. The hierarchy of controls in clause 8.1.2 (eliminate, substitute, engineer, administer, PPE) works better when the people who will live with the control help choose it.
4. Determining competence requirements, training needs, training, and evaluating training. Workers can tell you whether training worked, usually more accurately than a sign-in sheet.
5. Determining what needs to be monitored and measured and how. This is the participation counterpart to item 7 in the consultation list.
6. Investigating incidents and nonconformities and determining corrective actions. Clause 10.2 requires workers' participation in incident investigation. An investigation run entirely by supervisors and safety staff, with the injured worker interviewed but not involved, will draw attention.

Note 3 clarifies that emphasizing non-managerial workers is meant to apply to people carrying out the work activities, but is not intended to exclude, for example, managers who are affected by work activities or other organizational factors. So a front-line supervisor who is exposed to the same hazards belongs in the conversation too.

Note 4 is short and practical: providing training at no cost to workers, and during working hours where possible, can remove significant barriers to participation. If your workers must attend safety training on their own time or pay for their own certifications, you have probably created exactly the barrier 5.4 c) tells you to remove.

How do you build a clause 5.4 process that works?

Here is the sequence I use with organizations that are starting from scratch or repairing a weak process.

Step 1: Map your workers. List every group working under your control: employees by shift and site, agency workers, contractors, visitors who perform tasks, and any remote or mobile workers. Note language, literacy, and access constraints.

Step 2: Ask workers how they want to be involved. This satisfies 5.4 e) 1 and it changes the quality of everything that follows. A short facilitated session per area, or a brief survey, is enough to start.

Step 3: Choose mechanisms. Common ones are a joint health and safety committee, elected safety representatives, toolbox talks, task-level job hazard analysis, near-miss reporting, suggestion systems, and worker-led inspections. Match the mechanism to the topic. A committee suits policy review. A crew huddle suits a change in a task.

Step 4: Protect time and remove barriers. Schedule participation inside paid hours, translate materials, offer anonymous reporting where trust is low, and write down a no-reprisal commitment that management actually enforces.

Step 5: Link the process to clauses that use it. Hazard identification (6.1.2), change management (8.1.3), procurement and contractors (8.1.4), incident investigation (10.2), internal audit (9.2), and management review (9.3) each need a defined point where worker input enters. Our post on management of change shows one such connection in practice.

Step 6: Close the loop. For every input, record who received it, what was decided, and how the worker was told. Feeding back results is what turns a suggestion box into a participation process.

Step 7: Review effectiveness. Management review under clause 9.3 should consider the results of consultation and participation, and the internal audit should test whether the process is working in practice, not just whether it exists.

What documented information does clause 5.4 require?

The clause itself does not call for a specific document. Clause 7.5.1 requires documented information determined by the organization as necessary for the effectiveness of the system, and clause 7.4.1 requires you to retain documented information as evidence of communication where appropriate. In practice, auditors look for:

• A description of the consultation and participation process (procedure, flowchart, or section of the manual)
• Committee or representative terms of reference, where used
• Meeting minutes with attendee lists showing non-managerial participation
• Records of input received and the response given
• Evidence of worker involvement in specific events: risk assessments, investigations, procedure revisions
• Training and communication records showing time and resources were provided

For a fuller list of what a certification body expects across the whole system, see our required documents list.

What do auditors commonly find wrong with clause 5.4?

These are patterns I see repeatedly. This reflects my own experience, not frequency data, and the order is not a ranking.

A process that looks complete on paper can still fail an interview. Auditors are expected to speak with workers, and a worker who cannot describe how they raise a hazard, or what happened the last time they did, tells the auditor more than the procedure does.

How does clause 5.4 connect to the rest of ISO 45001?

Clause 5.4 sits in the leadership section, which tells you something about the intent: top management is responsible for making participation real. Clause 5.1 expects leaders to ensure the process exists and to protect workers from reprisals when they report. Clause 5.3 assigns roles, and 5.4 d) 3 requires consultation on those roles.

For a wider view of how clause 5.4 fits with planning, operation and improvement, our ISO 45001 implementation guide covers the full sequence.

Clause 5.4 checklist

Use this as a quick self-assessment.

• [ ] We have a defined process, not just a committee
• [ ] We have listed every category of worker under our control
• [ ] Workers helped design the consultation and participation mechanisms
• [ ] Time, training and resources are provided during paid hours
• [ ] Information is available in clear language and the languages used on site
• [ ] We have identified barriers and have a plan for each
• [ ] A no-reprisal commitment exists and has been tested in practice
• [ ] Non-managerial workers were consulted on policy, objectives, legal compliance, contractor controls, monitoring and the audit programme
• [ ] Non-managerial workers participate in hazard identification, control selection, training, monitoring and incident investigation
• [ ] Every input receives a recorded response
• [ ] Management review looks at how well consultation and participation are working

If several boxes are unticked, a sensible place to start is the feedback loop and the incident investigation role. In my experience those two changes do the most to shift how workers see the whole system.

I have come to think that the real test of clause 5.4 is simple: if you asked a night-shift operator how they influenced the last safety decision on their line, could they answer? If they could, your process is probably working. If they would look at you blankly, no amount of documentation will fix that.

If you want help assessing your own process against the clause, you can reach out through the Certify Consulting team or our contact page.

Jared Clark, JD, MBA, PMP, CMQ-OE, CQA, CPGP, RAC, Principal Consultant at Certify Consulting

Last updated: 2026-10-15