A near miss and a lost-time injury usually differ by luck, not by cause — the same missing guard, the same unsecured load, one step to the left. That is the whole argument for recording them: they are the only source of failure data a workplace gets without paying for it.
What it means in practice
- Reporting is internal, and voluntary in most jurisdictions. Some acts require notice of certain non-injury events — an explosion, a structural failure, an unplanned release — but a general near-miss duty is the exception.
- The counting rules are yours, which is exactly why they must be written down. A near-miss rate is only comparable to itself, and only if the definition did not drift.
- A rising number is usually good news. Reports go up when people trust the system. Reading the curve as a deterioration is the fastest way to close it.
- Anonymity buys volume, attribution buys follow-up. Most working systems offer both channels and publish what was fixed.
- Feed it into the same investigation process as injuries, with the same root-cause discipline and the same corrective-action tracking, or the reports become a suggestion box.
What it is not
It is not a first aid case. A first aid case is an injury, it is recorded under the first aid duty, and in several jurisdictions the record has to be retained for years. A near miss produced no injury at all.
It is not a hazard report. A hazard is a condition; a near miss is an event. Both belong in the system, and mixing them makes the numbers meaningless.
And it does not carry the weight it does in Italy. There, mancato infortunio reporting sits inside the D.Lgs. 231 organisational model and is scored in INAIL’s OT23 form, which turns it into a filing with a premium reduction attached. In Canada the same event has no external audience — which means the only reason to run the system is that it works.