The TRIR calculator above performs one piece of arithmetic that safety reporting has standardised on: it takes a count of recordable injury and illness cases, divides it by the hours actually worked, and scales the result to a fixed exposure base so that a small crew and a large one can be compared on the same axis. That base is 200,000 hours, and OSHA explains in a published standard interpretation on how incident rates are calculated that the figure represents 100 employees working 40 hours a week for 50 weeks a year.
Arb Digital builds free calculators that name the source of the convention they implement and stop where arithmetic stops. On this page the boundary matters more than usual. The calculator can divide accurately. It cannot tell you whether a particular case is recordable, because that determination is made under OSHA's recordkeeping regulation and by the people applying it to a specific set of facts — not by a web page and not by us.
What This TRIR Calculator Does
It reports four rates plus two exposure figures. The headline is the total recordable incident rate: every recordable case, scaled to the basis. The grid adds the DART rate, which counts only cases involving days away from work, restricted work activity or transfer to another job; the days-away case rate, a narrower cut still; a severity rate expressing total days away and restricted against the same basis; and the number of hours worked per recordable case, which is the same information turned inside out.
Below those sits a comparison against a benchmark rate you enter yourself. The page publishes no industry figures. Rates by industry are compiled and released by national statistics agencies from establishment-level reporting, they are revised, and quoting a stale one is worse than quoting none. Enter the current figure for your own NAICS code from the agency that publishes it and the tool will show the ratio; leave it at zero and the comparison is simply omitted.
Nothing here is adjacent to another tool on this site in the way that most of our calculators are, but two nearby pages are worth naming so you can see the boundary. The employee turnover rate calculator is a workforce metric with the same shape — a count over an exposure base — and no relationship at all to safety. The factor of safety calculator shares a word and nothing else: it is a structural margin between an applied load and a failure load, computed in advance of anything happening. TRIR is measured after the fact, and that distinction is the single most important thing to understand about it.
How to Use It
- Take the case counts from the summary of your OSHA Form 300A for the period, or from whatever equivalent log your jurisdiction requires. Enter total recordable cases first.
- Enter DART cases — those involving days away, restricted duty or job transfer — and then the narrower count of cases that involved days away from work. The second is a subset of the first.
- Enter total hours worked by all employees during the period. This is hours actually worked, not hours paid, so paid leave comes out and overtime goes in.
- Add total days away plus days restricted if you track them, to get a severity figure alongside the frequency figures.
- Enter the benchmark rate for your own industry classification, from the agency that publishes it, if you want the comparison. Leave it at zero if you do not have a current one.
The Formula and How It Is Calculated
Every rate on this page has the same structure:
Rate = (Number of cases × 200,000) ÷ Total hours worked by all employees
and the DART rate substitutes DART cases for total cases while leaving everything else identical. OSHA's recordkeeping requirements pages set out which employers must maintain the log, which cases go on it, and the forms that carry the annual summary.
Work through the defaults, which are a small contractor's year. Six recordable cases, two of them DART, one involving days away, against 350,000 hours worked. The TRIR is 6 × 200,000 ÷ 350,000 = 1,200,000 ÷ 350,000 = 3.43. The DART rate is 2 × 200,000 ÷ 350,000 = 1.14. The days-away case rate is 1 × 200,000 ÷ 350,000 = 0.57. With 45 combined days away and restricted, the severity rate is 45 × 200,000 ÷ 350,000 = 25.7 days per 200,000 hours.
The exposure figures come from the same two numbers. 350,000 hours divided by 6 cases is one recordable case every 58,333 hours worked. And 350,000 ÷ 2,000 is 175 full-time-equivalent workers, which is what the basis is really asking you to imagine: if this exposure were 175 people working a full year, roughly 3.4 recordable cases would appear per hundred of them.
What Counts as Recordable Is Not This Page’s Judgement
This deserves its own section because it is where the number is actually decided, and it is entirely outside the arithmetic. Whether an event goes on the log is governed by OSHA’s recordkeeping regulation, which sets out a general recording criterion, specific criteria for certain conditions, and a set of exceptions for things that happen at work but are not treated as work-related. Case determinations turn on facts — what treatment was given, whether the condition was significantly aggravated by work, whether an exception applies — and reasonable, well-trained people sometimes reach different conclusions on the same file.
The practical consequence is that the denominator of any TRIR comparison is more reliable than the numerator. Hours worked come out of a payroll system. Case counts come out of judgement calls, and judgement calls vary between employers, between sites within one employer, and between one recordkeeper and their successor. When a rate moves, the first question worth asking is whether the recording practice moved with it.
That variability also means the rate is manipulable in a way hours worked are not, and the pressure runs one direction. A site with a bonus tied to TRIR has a reason to look hard for a reason not to record. Nothing in this calculator can detect that.
TRIR Is a Lagging Indicator
A lagging indicator counts things that have already gone wrong. It tells you about outcomes in a period that is over, and it tells you nothing directly about the conditions that produced them or the conditions in place today. This is the most-repeated caution about TRIR and it is repeated because it keeps being ignored.
Three specific weaknesses follow from it. The first is statistical. Recordable cases are rare events, and rates computed on small exposures are dominated by noise. On 100,000 hours, a single extra case moves the TRIR by two whole points. Comparing one small site’s year against another’s, or one quarter against the last, is usually comparing random variation.
The second is that frequency and severity are different questions. A year with eight minor recordables and no lost time has a higher TRIR than a year with two cases, one of them catastrophic. Any organisation with serious-injury and fatality exposure has learned that the rate of small events is a poor proxy for the risk of a large one, which is why severity, potential-severity and precursor tracking sit alongside the rate rather than under it.
The third is that it says nothing about the work. Whether hazards are being identified, whether people report near misses without fear, whether a permit system is followed when nobody is watching — none of that appears in a count of injuries. Leading indicators exist to cover that ground: inspection close-out rates, training currency, observation quality, corrective-action ageing. A rate is a scoreboard. It is not a diagnosis, and treating a low one as evidence of a healthy safety culture is the classic error.
Comparing Rates Without the Right Benchmark Is Meaningless
Industries differ enormously in their baseline exposure, and a rate carries no meaning until it is placed against the right comparison group. Statistics agencies publish incident rates by NAICS industry code precisely so that a comparison can be made like for like. A rate that is unremarkable in one industry would be an outlier in another, and a company comparing itself against an all-industry average is learning almost nothing about its own performance.
Three rules keep a comparison honest. Use the most specific NAICS code that genuinely describes the work, not the parent code that flatters it. Use the same reference year for both figures, since published rates are revised and shift with reporting scope. And state the basis you used, because a rate computed on 1,000,000 hours is five times the same rate on 200,000 hours, and mixed-basis comparisons are a common and completely silent error — which is why the basis selector on this page reports which one it applied.
The other half of a fair comparison is the mix of work inside the number. A firm whose hours are mostly office-based will post a low combined rate however its field crews are doing, so separate rates per division are more informative than one company-wide figure.
Where the Hours Figure Goes Wrong
Because the denominator is usually the more trustworthy half, it is worth protecting. The most frequent error is using paid hours instead of hours worked. Holiday, vacation and sick pay are hours paid for time not worked, and including them inflates the denominator and deflates every rate on the page. The second is forgetting overtime, which pushes the rate the other way.
The third is scope mismatch: counting cases for a population whose hours are not all in the denominator, or vice versa. Temporary workers under an employer’s day-to-day supervision are a common source of this, since their injuries may be recordable by the host while their hours sit in an agency’s payroll. Whatever the answer is for your arrangement, the numerator and denominator have to describe the same population, or the ratio is not a rate at all.
Where hours are genuinely not tracked — small operations, salaried field staff — an estimate has to be built and documented, and its assumptions stated whenever the rate is quoted. Our absence rate calculator and Bradford factor calculator deal with the attendance side of the same payroll data, and the productivity calculator uses hours worked as its denominator too, so the definition you settle on is worth applying consistently across all of them.
Arb Digital designs and builds free interactive calculators that show their arithmetic, cite their sources and earn links because they are genuinely useful. Browse what we have already published, or tell us what your audience keeps searching for.
Browse the Free Tools Hub Talk to Arb DigitalCommon Mistakes to Avoid
- Using hours paid instead of hours worked. Paid leave is not exposure. Including it quietly lowers every rate on the page.
- Mixing rate bases. A figure computed on 1,000,000 hours is five times the same figure on 200,000. Always state which basis a rate uses before comparing it with anything.
- Comparing against an all-industry average. Rates are published by NAICS code because baseline exposure differs by industry. Without the right code the comparison teaches nothing.
- Reading a low rate as a safe operation. TRIR is a lagging count of recorded outcomes. It cannot see hazards, reporting culture, or the exposure that has not produced an injury yet.
- Treating quarter-to-quarter movement on small exposures as a trend. On low hour counts a single case swings the rate by whole points, and the swing is usually noise.
Related Free Tools From Arb Digital
For workforce metrics that share the count-over-exposure shape, the employee turnover rate calculator and the absence rate calculator work the same way on different data, and the Bradford factor calculator weights absence by frequency rather than duration. The productivity calculator uses the same hours-worked denominator for output. On the engineering side, the factor of safety calculator and the margin of safety calculator deal with design margins rather than recorded outcomes. Everything else is on the free online tools hub.
Frequently Asked Questions
Number of recordable cases multiplied by 200,000, divided by total hours worked by all employees in the period. OSHA explains in a published standard interpretation that the 200,000 figure represents 100 employees working 40 hours a week for 50 weeks a year, which is what makes rates comparable across employers of different sizes.
It is the annual hours of a notional 100 full-time workers. Scaling to that base means a rate can be read as cases per 100 full-time-equivalent workers per year, so a twelve-person shop and a twelve-thousand-person company land on the same axis. Some sectors report on a 1,000,000-hour base instead, which produces figures five times larger for identical performance.
TRIR counts every recordable case. The DART rate counts only the subset involving days away from work, restricted work activity or transfer to another job, so it is always the lower of the two. Both use the same denominator and the same 200,000-hour basis, and the gap between them says something about the severity mix of the cases behind the count.
That is defined by OSHA’s recordkeeping regulation, not by this calculator. The regulation sets a general recording criterion, specific criteria for certain conditions, and exceptions for events that occur at work but are not treated as work-related. Determinations turn on the facts of each case and are made by the employer applying the regulation, with the agency’s guidance as the reference.
The question cannot be answered in the abstract, because baseline exposure differs enormously by industry. A rate only becomes interpretable against the published figure for the same NAICS industry code, from the same reference year, on the same hour basis. This page holds no industry data and takes the benchmark as an input for that reason.
No. It is a lagging indicator: a count of recorded outcomes from a period that has ended. It cannot see the hazards present today, whether near misses are reported, or the potential severity behind the events that did occur. Organisations with serious-injury exposure track leading indicators and potential severity alongside the rate rather than relying on it.
Hours actually worked. Holiday, vacation and sick leave are paid but not worked, so including them inflates the denominator and understates every rate. Overtime is worked and must be included. The numerator and denominator also have to cover the same population, which is where arrangements involving temporary or contracted workers most often go wrong.
An estimate has to be built and documented, and its assumptions stated whenever the rate is quoted. A common approach is full-time-equivalent headcount multiplied by scheduled annual hours, adjusted for overtime and leave. Because the denominator is normally the more reliable half of the ratio, an undocumented estimate weakens the whole figure.
This page performs arithmetic on figures you enter. It is not a compliance assessment and it does not determine whether any case is recordable — that is defined by OSHA’s recordkeeping regulation and decided by the employer applying it. Incident rates are lagging indicators and say nothing on their own about the safety of current work. For recordkeeping obligations, case determinations or enforcement questions, consult OSHA or your jurisdiction’s occupational safety authority.
