When does a risk assessment actually need reviewing?

A lot of workplace risk assessments have a review date written at the bottom.

Quite often, that date is 12 months after the assessment was completed.

There’s nothing wrong with setting an annual review date. It gives the business a prompt to go back and check that the assessment still reflects the work being carried out.

The problem comes when that date becomes the only reason anyone looks at the assessment again.

A workplace can change considerably in 12 months.

New equipment might be introduced. A process might change. An accident could expose a weakness in the existing controls. Staff might start carrying out a task differently from the way it was originally assessed.

If any of that happens, waiting for the review date probably doesn’t make much sense.

Is an annual review actually a legal requirement?

Not as a general rule.

Regulation 3 of the Management of Health and Safety at Work Regulations 1999 requires an assessment to be reviewed where there is reason to suspect it is no longer valid, or where there has been a significant change in the matters to which it relates.

So the important question isn't:

“Is the review date due?”

It is:

“Does this assessment still accurately reflect the work and the risks?”

HSE guidance also makes clear that risk assessments should be kept up to date and reviewed when circumstances change.

A scheduled review remains useful. It just shouldn't stop an earlier review happening when something has changed.

The work has changed

This is probably the most obvious trigger.

Imagine a warehouse risk assessment was completed six months ago.

Since then:

• racking has been moved;
• the picking route has changed;
• a new type of handling equipment has been introduced; and
• employees are now moving products differently.

The document might still be only six months old, but parts of it could already be out of date.

A review should consider whether those changes have introduced different hazards, altered the level of risk or affected the controls already in place.

Ideally, significant changes should be considered before they are introduced rather than after the new process is already operating.

New equipment, substances or processes are introduced

A new piece of machinery can change more than just the task it is being used for.

It might introduce:

• new maintenance requirements;
• different manual handling risks;
• additional noise;
• vehicle or pedestrian interactions;
• new training requirements; or
• new isolation arrangements.

The same applies to new substances and changes in working methods.

Even if the overall job hasn't changed, the way people are exposed to risk might have.

That should prompt another look at the assessment.

There has been an accident or near miss

An incident doesn't automatically mean the original assessment was poor.

But it does give you new information.

If someone is injured carrying out a task that has already been assessed, it is worth asking whether the assessment properly reflected what was happening in practice.

Perhaps a hazard wasn't identified.

Perhaps the control looked suitable on paper but wasn't practical.

Or perhaps the control was perfectly reasonable but wasn't actually being followed.

Near misses can be just as useful.

If several people nearly slip in the same area, repeatedly struggle with the same load or report the same unsafe condition, that is information worth feeding back into the assessment process.

HSE guidance specifically identifies accidents, near misses and problems raised by workers as reasons to consider reviewing controls.

The controls aren't working

Sometimes nothing obvious has changed, but the assessment still isn't doing what it needs to do.

Imagine a manual handling assessment says employees should use mechanical assistance for a particular product.

During an inspection, you discover people are regularly lifting it manually instead.

The useful response isn't simply:

“Remind staff to follow the risk assessment.”

Find out why.

Is the equipment unavailable?

Does it take too long to access?

Is there enough space to use it?

Was the control unrealistic in the first place?

A review needs to look at how the work is actually carried out, not just whether the paperwork still looks correct.

People doing the work raise concerns

The people carrying out a task often notice changes before anybody reviewing the paperwork does.

They know when equipment has become awkward to use.

They know when a process has gradually changed.

And they usually know which controls work well and which ones are being worked around.

That doesn't mean every suggestion automatically requires the assessment to be rewritten.

It does mean employee feedback should be considered when deciding whether the assessment still represents the real task.

A review carried out entirely from behind a desk can easily miss that.

What should a review actually involve?

Changing the review date and adding a signature isn't much of a review.

A meaningful review should ask:

• Is the activity still being carried out in the same way?
• Are the hazards still accurate?
• Has anyone new or different become exposed?
• Are the existing controls actually in place?
• Are those controls effective?
• Have there been incidents, near misses or concerns?
• Has new equipment, machinery or a substance been introduced?
• Does anything need changing?

If nothing has changed and the controls remain suitable, that's fine.

Record that the assessment has been reviewed and remains appropriate.

You don't need to rewrite a perfectly good assessment just so the document looks different.

Don't let the date manage the assessment

Review dates are useful administrative controls.

They help prevent risk assessments from being forgotten completely.

But the calendar doesn't know that you've changed the warehouse layout.

It doesn't know that a new machine arrived last Tuesday.

And it doesn't know somebody nearly had an accident yesterday.

The assessment should change when the risk changes, not simply when a date arrives.

A useful question to ask during any review is:

If we assessed this job from scratch today, knowing what we know now, would we reach the same conclusion?

If the answer is no, the assessment probably needs updating.

Keeping reviews visible

This gets more difficult when an organisation has multiple sites, projects or large numbers of assessments.

Knowing which assessments exist, when they were last reviewed and which ones may need attention can quickly become an administrative job in itself.

That's where a system such as SafeChain can help by keeping risk assessments and their review information in one place.

But the software isn't the review.

Someone still needs to look at the work, speak to the people involved and decide whether the controls remain suitable.

Technology can provide the visibility.

The judgement still comes from the people managing the risk.

Good Safety Needs Good Visibility.