Near miss: meaning in UK construction

A near miss is an unplanned event that did not result in injury, illness or damage, but realistically could have. A load swinging where someone had been standing a minute earlier. A scaffold board that slipped. A reversing vehicle that stopped short.

The value of a near miss is that it hands you the information an accident would have given you, without the accident. It is the cheapest safety intelligence a business will ever get, and most of it goes unreported.

Why reporting fails

Three reasons, and all of them are fixable.

It is too slow. If reporting means finding a form, filling it in and getting it to the office, it will not happen mid-shift. Reporting from a phone, at the point it happened, with a photo attached, is the difference between a report and a shrug.

It feels like blame. Where a report leads to someone being disciplined, reporting stops. Near miss systems only work when they are explicitly about the circumstances rather than the person.

Nothing happens. This is the one that kills mature schemes. People report for a while, see no change, and conclude it is pointless. Telling the site what was fixed as a result of last month’s reports is what keeps the flow going.

What to do with them

The point is to find the cause and remove it, not to record that it happened. A report closed with a note saying operatives were reminded to take care has changed nothing, because a reminder sits near the bottom of the hierarchy of control. One that ends in edge protection being extended, a route being changed, or a delivery sequence being altered has actually removed the event.

Patterns matter more than individual reports. Three near misses at the same gate, or four involving the same subcontractor, are telling you something a single report cannot. That is where near miss data feeds usefully into audits and the next toolbox talk.

The line with RIDDOR

Most near misses are internal information. Some are not: RIDDOR requires certain specified dangerous occurrences to be reported to the enforcing authority even where nobody was hurt, such as a scaffold collapse or the accidental release of a biological agent. Knowing which of your near misses cross that line is part of running the scheme properly.

Making the report take a minute from a phone on site, with the project and photos attached, is what produces the volume that makes the data useful. That is what Sync’s incident reporting is built for.

Common questions

Do near misses have to be reported to the HSE?

Generally no. RIDDOR requires certain specified dangerous occurrences to be reported even where nobody was injured, so some events that feel like near misses are legally reportable. Most are not, and are reported internally so the cause can be dealt with.

Why bother reporting if nobody was hurt?

Because the same set of circumstances will recur, and next time the outcome may be different. A near miss gives you the information an injury would have given you, without the injury, the investigation, the lost time or the claim. It is the cheapest safety data available.

How do you get people to actually report?

Make it quick, make it blameless, and close the loop. If reporting takes ten minutes of paperwork, nobody does it. If it results in someone being blamed, nobody does it twice. If nothing visibly changes afterwards, people conclude it is pointless. Telling the site what was fixed as a result is what sustains reporting.

What should be done with a report?

Find the cause rather than the culprit, then fix it as far up the hierarchy of control as you can. A report that ends in reminding people to be careful has changed nothing. One that ends in a physical change, a sequence change or a design change has removed the event.

Near miss