A plant with 40 employees has one lost-time injury in March. By April, its TRIR has jumped from 0 to something that looks alarming on a dashboard — even though nothing else changed. That's the trouble with the Total Recordable Incident Rate: it's the most quoted safety number in the building, but almost nobody outside the safety office knows what a "good" one actually looks like, or how fast it can swing. This piece walks through the TRIR calculation OSHA-covered employers use, works a real example line by line, and shows where the metric quietly misleads people who read it without context.
What is TRIR and what does it actually measure?
TRIR stands for Total Recordable Incident Rate. It's a standardised way to compare safety performance across companies of different sizes, using data pulled straight from your OSHA 300 Log. The formula is:
TRIR = (Number of OSHA recordable cases × 200,000) ÷ Total hours worked by all employees
Every piece of that formula does a specific job, so it's worth breaking apart.
Number of OSHA recordable cases. This is the numerator, and it only counts cases that clear the recordability test in 1904.7: a death, days away from work, restricted work or job transfer, medical treatment beyond first aid, loss of consciousness, or a significant injury or illness diagnosed by a physician or licensed health care professional. First aid cases don't count. If you're not sure whether a specific case belongs in that number, Job13's free recordability check runs the test against the facts and gives you a clear answer — or flags it as needing review when the regulation genuinely leaves room for judgement, rather than guessing. Getting this number wrong at the source poisons every rate you calculate from it.
200,000. This is a fixed constant, not something you calculate. It represents 100 employees working 40 hours a week, 50 weeks a year — a standard "100 full-time workers, one year" baseline. It exists purely to standardise the rate, so a 40-person shop and a 4,000-person plant produce numbers you can put side by side.
Total hours worked. This is the denominator, and it's where most calculation errors happen. It means actual hours worked by all employees during the period — including overtime — not hours scheduled or hours paid. Vacation, sick leave and holiday time that wasn't actually worked doesn't count.
What is a good TRIR?
There's no single number that counts as "good" across every industry — a warehouse and an accounting firm are never going to sit on the same scale, and pretending otherwise is how TRIR gets misused. What you can say directionally: office-based and low-hazard service work tends to cluster at the low end, general manufacturing and warehousing sit in the middle, and higher-risk physical trades — heavy construction, certain manufacturing sub-sectors, waste handling — trend meaningfully higher. OSHA and the Bureau of Labor Statistics publish industry-average rates by NAICS code that let you benchmark against your specific peer group rather than a generic number pulled from a blog post. Comparing your TRIR to those published averages for your actual industry is far more useful than comparing it to a flat "under 1.0 is good" rule of thumb that a warehouse and a call centre can't both meet fairly.
A fully worked TRIR calculation
Here's a mid-sized manufacturer, straightforward numbers, so you can see exactly where each figure comes from.
| Step | Value | Where it comes from |
|---|---|---|
| Recordable cases (from the 300 Log) | 6 | Count of cases meeting 1904.7 criteria for the year |
| Constant | 200,000 | Fixed — represents 100 employees × 40 hrs/week × 50 weeks |
| Numerator: 6 × 200,000 | 1,200,000 | Cases multiplied by the constant |
| Total hours worked | 480,000 | 240 employees × ~2,000 actual hours worked each, from payroll/timekeeping |
| TRIR: 1,200,000 ÷ 480,000 | 2.5 | Final rate |
Worked example: a 240-employee manufacturer with 6 recordable cases and 480,000 hours worked in the year produces a TRIR of 2.5.
Swap in your own case count and hours and the mechanics don't change. The formula never varies — only the two real inputs do, and both of those inputs live on your 300 Log and your timekeeping system, not in the formula itself. If you'd rather not build the spreadsheet from scratch, Job13's TRIR and DART calculator does this arithmetic for you and holds onto the same recordable-case data your 300 Log already tracks.
How is DART rate different from TRIR?
DART stands for Days Away, Restricted duty, or Transfer — and it's a narrower cut of the same data. The formula is identical in shape:
DART = (Cases involving days away, restricted duty, or transfer × 200,000) ÷ Total hours worked
The difference is the numerator. TRIR counts every recordable case, including ones resolved with medical treatment alone. DART only counts the more severe subset — cases where someone actually missed work, worked restricted duty, or transferred to a different job. A company can have a TRIR of 2.5 and a DART rate closer to 1.0, because several of its recordable cases were treated and closed without time away. We cover the mechanics in more depth in our DART rate explainer, and both rates sit side by side in the rate calculator.
Why one incident can swing a small company's TRIR wildly
Here's where TRIR stops being a neutral number and starts needing a health warning, especially for smaller employers.
Picture a 25-person electrical contracting firm. It works roughly 50,000 hours a year — 25 employees at around 2,000 hours each. For the first ten months of the year, it has zero recordable cases. TRIR sits at 0. Then, in November, an apprentice falls from a ladder and breaks a wrist: a recordable case with days away from work.
Run the math on that single case: (1 × 200,000) ÷ 50,000 = 4.0. One incident took the firm from a perfect 0 to a rate that would flag it as elevated against most industry benchmarks — and it happened in a single afternoon, on a single job.
Now compare that to a 2,000-person contractor working four million hours a year. If that company has the same one incident, its TRIR barely moves: (1 × 200,000) ÷ 4,000,000 = 0.05. Same event, same severity, wildly different signal.
One EHS manager we spoke with put it this way: "I've had board members ask why our TRIR tripled in a quarter, and the honest answer was one guy sprained an ankle. At our headcount, that's what the math does. It doesn't mean the whole safety programme fell apart overnight." That's not a flaw in the arithmetic — the formula is doing exactly what it's designed to do. The flaw is in reading a small company's single-month or single-quarter TRIR as if it carries the same statistical weight as a large company's annual rate. It doesn't. Smaller denominators mean more volatility, full stop.
Common pitfalls in TRIR calculations
Most bad TRIR numbers aren't the result of bad safety performance — they're the result of bad inputs. Watch for these:
- Using scheduled or paid hours instead of actual hours worked. Payroll systems often report hours paid, which includes vacation and sick time. TRIR wants hours actually worked. Using paid hours inflates the denominator and understates the rate, making performance look better than it is.
- Forgetting contractor and temp-staff hours. If your company supervises the day-to-day work of contract or temporary staff, their hours and their recordable cases generally belong in your count too. Leaving them out shrinks both numbers and distorts the rate — sometimes in your favour, sometimes not.
- Missing recordable cases at the source. If a case that should have gone on the 300 Log never made it there, your TRIR is wrong before you even divide. This is the single most common failure point, and it's worth running borderline cases through a proper check — see our recordability criteria checklist — rather than guessing at intake.
- Comparing TRIR across very different industries. A 1.5 in office administration and a 1.5 in structural steel erection do not mean the same thing. Benchmark against your own NAICS peer group using published OSHA or BLS data, not against a number from an unrelated sector.
- Reading short time windows as trends. A single quarter's TRIR, especially at a smaller site, can move sharply on one case. Look at rolling 12-month figures before drawing conclusions about whether performance is actually improving or declining.
Where TRIR breaks down as a standalone metric
TRIR was built to standardise, and it does that job well at scale. Where it struggles is exactly the small-employer scenario above, and a second, quieter problem: TRIR treats a broken wrist and a case that needed six stitches as equally weighted "recordable" events, provided both cleared the 1904.7 bar. It doesn't distinguish severity beyond the DART cut. A company chasing a lower TRIR number alone can end up under-recording borderline cases rather than genuinely reducing risk — which is exactly the kind of pressure that leads to disputes over first aid versus medical treatment. Our guide on first aid vs. medical treatment covers where that line actually sits under the rule.
None of this makes TRIR useless. It makes it a metric that needs context: your own trend over time, your genuine industry peer group, and an honest read of your headcount before anyone reacts to a single quarter's swing.
Get your TRIR calculated correctly
Getting TRIR right starts with getting recordability right, case by case, before it ever hits the log. Job13's free recordability check tells you plainly whether a specific case belongs on the 300 Log — or flags it for review when the facts genuinely require judgement — and paying users can keep the 300 Log, the 300A summary and the ITA export in one place. Once your case data is right, run the numbers through Job13's TRIR and DART calculator to see your rate instantly, with no spreadsheet formulas to get wrong.
This is general information about the 29 CFR Part 1904 recordkeeping rule, not legal advice.
Frequently asked questions
What is the TRIR formula?
TRIR equals the number of OSHA recordable cases multiplied by 200,000, divided by the total hours worked by all employees during the period. The 200,000 figure is a fixed constant representing 100 full-time employees working a standard year, and it exists to make companies of different sizes comparable on the same scale.
What hours count toward the TRIR calculation?
Total hours worked means actual hours employees worked, including overtime. It excludes vacation, sick leave and holiday time that wasn't actually worked. Using scheduled or paid hours instead of actual worked hours is one of the most common errors in a TRIR calculation, and it usually makes the rate look better than it really is.
What counts as a recordable case in the TRIR numerator?
A recordable case is one that meets the criteria in 1904.7: a death, days away from work, restricted work or job transfer, medical treatment beyond first aid, loss of consciousness, or a significant injury or illness diagnosed by a physician or licensed health care professional. First aid alone does not make a case recordable. Job13's recordability check applies this test to a specific case and returns a clear answer or a "needs review" flag when the facts leave real judgement involved.
How is TRIR different from DART rate?
Both use the same formula shape and the same 200,000 constant, but they count different things in the numerator. TRIR counts every recordable case. DART counts only the more severe subset — cases involving days away from work, restricted duty, or job transfer. DART is always equal to or lower than TRIR for the same period, because it's a subset of the same case pool.
Why is my small company's TRIR so volatile?
Because the denominator — total hours worked — is small for a small company, one recordable case moves the rate a lot more than it would at a larger employer. A 25-person firm with one lost-time injury can see its TRIR jump from 0 to 4.0 or higher in a single incident, while the same injury barely registers at a company with millions of annual work hours. This is a real, well-understood limit of the metric, not a sign the underlying formula is broken — it just means small companies should read short-term TRIR swings with caution and look at longer trends before drawing conclusions.
