Most workplaces track injuries. Fewer track them in a way that actually tells them something useful.
The difference between a safety program that improves over time and one that just reacts to incidents usually comes down to measurement. Not how much you measure, but whether you're measuring the right things — and whether the numbers mean anything to the people reading them.
Here are the five metrics that give you the clearest picture of workplace safety performance, what they measure, and how to calculate each one.
1. LTIFR — Lost Time Injury Frequency Rate
LTIFR is the most widely used safety metric globally. It tells you how many lost time injuries occurred per million hours worked.
A lost time injury is any work-related injury or illness that causes a worker to miss at least one scheduled shift.
The formula:
LTIFR = (Number of lost time injuries x 1,000,000) / Hours worked
So if your workforce logged 500,000 hours over a period and had 3 lost time injuries, your LTIFR is 6.
The million-hour multiplier is just a convention — it makes the number readable regardless of workforce size. A company with 20 workers and one with 2,000 workers can compare their LTIFR directly.
What counts as a good LTIFR varies by industry. Construction and mining run higher than office environments. The number only becomes meaningful when you track it over time or benchmark it against your sector's average.
You can use a free LTIFR calculator to run the numbers quickly, or do it manually using the formula above.
2. TRIR — Total Recordable Incident Rate
LTIFR only counts injuries serious enough to cause lost time. TRIR casts a wider net.
It includes any work-related injury or illness that results in:
- Lost time (days away from work)
- Restricted work or job transfer
- Medical treatment beyond first aid
The formula:
TRIR = (Number of recordable incidents x 200,000) / Hours worked
The 200,000 figure represents 100 employees working 40 hours a week for 50 weeks — a standard US OSHA baseline. It normalises the rate per 100 full-time workers.
TRIR is a broader indicator of safety performance than LTIFR. If your LTIFR looks fine but your TRIR is climbing, it usually means you're accumulating a lot of minor recordable injuries that haven't yet caused lost time. That's a warning sign worth acting on early.
3. AFR — Accident Frequency Rate
AFR is similar to LTIFR but uses 100,000 hours as the denominator instead of 1,000,000. It's more common in the UK and parts of Europe, though you'll see it used interchangeably with LTIFR in some organisations.
The formula:
AFR = (Number of accidents x 100,000) / Hours worked
The main thing to know about AFR is the denominator difference. If someone quotes you an AFR, make sure you understand whether they mean per 100,000 or per 1,000,000 hours worked. Mixing them up makes benchmarking meaningless.
4. DART Rate — Days Away, Restricted, or Transferred
DART rate comes from OSHA's recordkeeping framework and measures the rate of injuries and illnesses that caused workers to miss work, work restricted duties, or transfer to another job.
The formula:
DART Rate = (DART cases x 200,000) / Hours worked
The difference between DART and TRIR is that DART excludes cases that only required medical treatment. A worker who needed stitches and came back to their normal job the next day is a TRIR case but not a DART case.
DART rate is a sharper measure of incident severity. Two companies might have the same TRIR but very different DART rates — which tells you something real about the nature of the injuries each is experiencing.
5. Near Miss Rate
Near miss rate is the metric most organisations track worst, and arguably the one that predicts future injuries most reliably.
A near miss is an unplanned event that didn't result in injury or damage but had the potential to. A scaffold board that falls but misses the worker below. A chemical spill that gets contained before anyone is exposed. A forklift that clips a shelf but doesn't tip.
There's no universal formula for near miss rate, but a common approach is:
Near Miss Rate = (Number of near misses x 1,000,000) / Hours worked
The challenge with near miss data isn't the calculation — it's the reporting. Workers under-report near misses when they fear blame, don't see the point, or think it creates paperwork for no benefit. If your near miss rate is very low, that's often a sign of a reporting culture problem, not a safe workplace.
Organisations with strong safety cultures typically report far more near misses than organisations with weak ones. Counter-intuitive, but consistent.
How to use these metrics together
No single metric gives you the full picture on its own.
LTIFR tells you about serious injuries. TRIR tells you about the broader volume of recordable events. DART rate tells you about severity. Near miss rate tells you what's brewing before anything gets recorded officially.
Track all five over rolling 12-month periods rather than calendar years — it smooths out seasonal variation and gives you a cleaner trend line. And always relate the numbers back to hours worked, not headcount. Headcount changes. Hours worked is the honest denominator.
But here's the thing metrics can't do on their own: tell you what to fix.
Once your numbers flag a problem — a rising TRIR, a spike in near misses in a specific area — you need a structured way to respond. That's where the hierarchy of controls comes in. It's a framework for ranking corrective actions from most to least effective, and it stops teams from defaulting to the easiest fix (usually more training or more signage) rather than the right one.
The goal isn't to hit a target number. It's to understand whether things are getting better or worse, act on what the data is telling you, and make sure the actions you take actually address the root cause.
