TRIFR and LTIFR Explained: The Two Safety Numbers Clients Ask For

It is 4pm on a Thursday and the prequalification return for a $60M school project is due at noon tomorrow. The HSEQ lead has filled in everything except one table: TRIFR and LTIFR for each of the last three financial years, total hours worked, and a tick box asking whether subcontractor hours are included.

She has three spreadsheets. The one from the old safety system says the TRIFR was 8.2 last year. The monthly board report says 11.6. The spreadsheet the previous HSEQ manager kept says 21.4. All three used the same formula. None of them agree, and she has until the morning to work out which one she can defend.

What is TRIFR?

TRIFR stands for total recordable injury frequency rate. It is the number of recordable injuries per million hours worked over a period, and it is the headline safety number in most Australian construction tenders, prequalification questionnaires and board packs.

A recordable injury is generally any work-related injury more serious than first aid. In practice that means the sum of four categories:

• Fatalities

• Lost time injuries (LTIs): the worker lost at least one full shift, most commonly counted from the shift after the injury

• Medical treatment injuries (MTIs): treatment beyond first aid by a doctor or other health professional, with no time lost

• Restricted work injuries (RWIs): the worker came back on modified or alternative duties

First aid injuries are excluded. So are near misses, which are counted separately and matter a great deal, as we come back to below.

What is LTIFR?

LTIFR is the lost time injury frequency rate: lost time injuries per million hours worked. It is a subset of TRIFR, counting only the injuries serious enough to cost a shift.

LTIFR is the older of the two measures and the one Safe Work Australia historically benchmarked. Its data is now published as the workers’ compensation injury frequency rate (WCIFR), defined as the number of claims per million hours worked, and filtered to lost time claims for the LTIFR equivalent.[1]

Two related rates turn up on the same forms. MTIFR counts medical treatment injuries only. AIFR, the all injury frequency rate, adds first aid cases to the recordables, which makes it the most sensitive and least comparable of the set.

How do you calculate TRIFR and LTIFR?

Every rate follows the same shape: injuries in the period, divided by hours worked in the period, multiplied by one million.

A worked example

A mid-tier builder has one financial year of data. Direct employees worked 240,000 hours. Subcontractor workers on its sites worked a further 360,000 hours. Over the year there were 2 LTIs, 5 MTIs, 2 RWIs and no fatalities, across both groups.

That is the correct answer. The next section is about the two wrong ones, which are more common.

9.3

Serious workers’ compensation claims per million hours worked in construction, against 6.8 across all industries (2023-24, preliminary).

Safe Work Australia, Key WHS Statistics 2025

Why do TRIFR numbers disagree?

The formula is not where TRIFR goes wrong. The inputs are, and the most common failure is a numerator and denominator that describe different workforces.

Take the worked example above. If the injury register counts every injury on site, subcontractors included, but the hours figure comes from payroll and covers direct employees only, the calculation becomes 9 × 1,000,000 ÷ 240,000. The TRIFR jumps to 37.5.

Flip it. Include all 600,000 site hours, but count only the three injuries that happened to direct employees, and the TRIFR falls to 5.0. It looks excellent, and it is wrong in the direction nobody questions.

The other quiet movers:

• The LTI definition. Whether the day of injury counts, and whether a worker on restricted duties is an LTI or an RWI, changes the number more than most safety programmes do. Safe Work Australia’s own claims data defines one day or shift lost as three hours.[1]

• Estimated hours. Headcount multiplied by a standard week is not hours worked. Construction runs on overtime, short weeks and wet weather days.

• Reclassification. An MTI quietly recorded as first aid disappears from the numerator. So does an LTI where the worker is offered a desk for a week.

• Small denominators. A small builder with 150,000 hours a year moves its TRIFR by 6.7 with a single injury.

Per million hours or per 200,000 hours?

Australia, the UK and most of the resources sector use one million hours. The United States uses 200,000, which OSHA explains is the hours that 100 employees working 40 hours a week, 50 weeks a year would work.[2]

The conversion is a factor of five, and it is the single most common error on forms filled in by multinational head contractors and international clients. A US TRIR of 3.0 is the same safety performance as an Australian TRIFR of 15.0. Put them side by side without converting and one company looks five times safer than the other.

Recordable definitions also differ between the two systems, so even a correctly converted figure is indicative rather than exact. OSHA’s own guidance notes that a single injury has a much greater effect on the rate in small establishments than in large ones.[2]

What is a good TRIFR?

This is the question everybody asks, and the honest answer is uncomfortable. There is no published national TRIFR benchmark in Australia, for construction or any other industry. Safe Work Australia benchmarks claims-based frequency rates, notes that you will not find TRIFR in the injury survey analysis it points to, and states that it does not provide other benchmarks such as MTIFR.[1]

A tender response claiming a TRIFR “well below the industry average” is comparing against a private dataset, an association survey, or a US figure on a different basis.

For LTIFR, the Safe Work Australia claims data is a legitimate reference point, with a caveat that runs in one direction. It counts accepted workers’ compensation claims, not all lost time injuries, and Safe Work Australia says so plainly.

“WCIFR is likely to be an underestimate of all lost time injuries due to some workers being ineligible for workers compensation and claims not being made for minor injuries.”

Safe Work Australia, Understanding Workers’ Compensation Injury Frequency Rates[1]

The ABS work-related injuries survey puts a number on that gap: only 31% of workers who experienced a work-related injury or illness in 2021-22 received workers’ compensation for it.[4] Your internal LTIFR counts more than the national figure does, so a like-for-like comparison flatters the benchmark.

The useful comparison for TRIFR is your own trend, over three to five years, by project and by trade.

31%

Share of Australian workers injured at work in 2021-22 who received workers’ compensation. Claims-based benchmarks miss the rest.

Australian Bureau of Statistics

Why TRIFR is not enough on its own

TRIFR and LTIFR are lagging indicators. They count harm that has already happened, and they reward a site where nobody reports anything as generously as a site where nothing happens.

A falling TRIFR with a falling near-miss reporting rate is not good news. It is usually a reporting culture problem that the frequency rate cannot see. Safe Work Australia’s guidance on WHS data reporting is explicit: do not rely on any one indicator such as injury rates, because it can miss important attributes of safety culture.[6]

The leading indicators that give TRIFR its context are the ones that describe what the business is actually doing:

• Near misses and hazard observations reported per 100,000 hours

• Corrective actions closed on time, with evidence

• Scheduled inspections and site walks completed

• Toolbox talk and prestart attendance

• High-risk permits open, and permits closed out properly

What good looks like

Teams that answer a prequalification table in ten minutes are not better at arithmetic. Their records make the arithmetic trivial.

• Hours worked come from sign-in and timesheet records for everyone on site, subcontractors included, not from headcount estimates

• Every injury is classified at the time (LTI, MTI, RWI or first aid) against one written definition used on every project

• Reclassifications are logged with a reason and an approver, so nobody can quietly move an MTI to first aid

• The rate can be cut by project, by trade and by period without rebuilding a spreadsheet

• TRIFR sits next to near-miss rates and corrective action close-out, so a falling number can be read properly

• The tick box asking whether subcontractor hours are included has a true answer, and the same answer every year

The Australian context

Construction carries more than its share of harm. According to Safe Work Australia’s key WHS statistics, in 2023-24 it accounted for 17,600 serious claims, 12.0% of the national total, at a frequency rate of 9.3 per million hours against 6.8 for all industries, with a median of 8.4 weeks lost per claim.[3]

For head contractors working on federally funded projects, these numbers are not only for tenders. Companies accredited under the Australian Government WHS Accreditation Scheme must report lost time injuries to the Office of the Federal Safety Commissioner on projects valued at $4 million or more, and medically treated injuries on Scheme projects, each within two weeks. They also submit biannual reports covering WHS performance indicators.[5]

That makes the injury classification a compliance record, not just a statistic. An LTI recorded as an MTI changes what gets reported to a federal regulator. If the incident also meets the notifiable threshold, the separate state reporting duty covered in notifiable incidents in construction applies on top.

Where Plexa fits

Plexa’s Executive WHS Safety Dashboard, one of the ready-made Workbench apps, shows days since last LTI, TRIFR, LTIFR, near misses and active high-risk permits in one live view. The frequency rates sit beside the leading indicators rather than in a separate report.

The inputs live in the same platform. Incidents and injuries are recorded on site with the affected person, severity and response. Sign-in and sign-out history is held against each worker, subcontractors included, and timesheets capture clock-in and clock-out, as covered in tracking construction labour cost. Because the numerator and the denominator describe the same workforce, the rate stops depending on which spreadsheet you open.

For a quick number today, Plexa’s free LTIFR and TRIFR calculator works one out and benchmarks LTIFR against Safe Work Australia’s published construction data.

Back at 4pm on Thursday, the HSEQ lead does not have to choose between 8.2, 11.6 and 21.4. She filters the dashboard by financial year, confirms subcontractor hours are in, ticks the box, and has the form back to the estimator before five.

Frequently asked questions

How do you calculate TRIFR? Add up fatalities, lost time injuries, medical treatment injuries and restricted work injuries for the period. Multiply by 1,000,000, then divide by total hours worked in the same period. Nine recordable injuries over 600,000 hours gives a TRIFR of 15.0.

What injuries count towards TRIFR? Fatalities, lost time injuries, medical treatment injuries and restricted work injuries. First aid injuries and near misses are excluded, although near misses should be tracked separately as a leading indicator.

What is the difference between TRIFR and LTIFR? LTIFR counts only lost time injuries, where at least one full shift was lost. TRIFR counts every recordable injury, including medical treatment and restricted work cases, so it is always equal to or higher than LTIFR for the same period.

What is a good TRIFR in construction? Australia has no published national TRIFR benchmark. Safe Work Australia benchmarks claims-based frequency rates and does not publish TRIFR.[1] Compare TRIFR against your own three to five year trend and across your own projects, and treat any quoted “industry average” TRIFR with caution.

Should subcontractor hours be included in TRIFR? If subcontractor injuries are in the numerator, subcontractor hours must be in the denominator. Most principal contractors report on the whole site workforce. Mixing the two produces a rate that is either inflated or flattered, and it is the most common reason TRIFR figures disagree.

Is TRIFR per million hours or per 200,000 hours? Australia uses one million hours. The US OSHA convention uses 200,000 hours, the annual hours of 100 full-time employees.[2] Multiply a US rate by five to put it on the Australian basis, and expect some remaining difference because recordable definitions are not identical.

Related reading

The wider set of duties these numbers sit inside is covered in what HSEQ actually means on a construction site. Why safety numbers in a monthly PDF arrive too late to act on is the subject of construction reporting. And the sign-in record that gives you an honest hours figure is explained in seven ways construction teams sign in to site.

If you want to see TRIFR, LTIFR and the leading indicators live across your projects, book a 30-minute demo with the Plexa team.