TRIR, DART, and EMR: Safety Metrics Explained with Calculation Examples
If a contractor prequalification form asks for your TRIR and DART rate, or an insurer quotes a premium tied to your EMR, you need to know exactly how those numbers are produced — and what they say about your program. Getting a formula wrong understates or inflates your performance, and either error costs you credibility in an audit or a bid. This article walks through the three metrics that decide how the outside world judges your safety record, with the formulas, worked calculations, and current benchmarks you need to interpret them.
The numbers are not interchangeable. TRIR and DART measure recorded injury frequency; EMR measures the cost your claims history imposes on your insurer. A company can post a respectable TRIR and still carry a punishing EMR. Understanding what each one actually counts is the difference between managing a metric and managing the risk underneath it.
Track every recordable in one place
WhyTrace Plus logs incidents, classifies recordable cases, and tracks the hours and cases behind your TRIR and DART rates — so the numbers on your bid documents come from your own data, not a year-end scramble. See how WhyTrace Plus tracks incident metrics →
TRIR: The Total Recordable Incident Rate
TRIR (Total Recordable Incident Rate) is the number of OSHA-recordable injuries and illnesses your organization records per 100 full-time workers over a year. It is the most widely cited safety frequency metric and the one most often requested on prequalification forms.
The formula uses a fixed 200,000-hour multiplier:
TRIR = (Number of recordable cases × 200,000) ÷ Total hours worked
The 200,000 figure represents the hours 100 employees would work in a year — 100 workers × 40 hours × 50 weeks. Normalizing to that base lets a 30-person shop and a 3,000-person plant compare rates on the same scale (source: Creative Safety Supply, OSHA Incident Rates, accessed 2026).
What counts as recordable. Under OSHA 29 CFR 1904, a case is recordable if it involves death, days away from work, restricted work or job transfer, medical treatment beyond first aid, loss of consciousness, or a significant diagnosed injury or illness. First-aid-only cases are not recordable and do not enter the TRIR numerator.
Worked example. A facility records 5 recordable cases and its employees worked 250,000 hours in the year.
| Input | Value |
|---|---|
| Recordable cases | 5 |
| Hours worked | 250,000 |
| Calculation | (5 × 200,000) ÷ 250,000 |
| TRIR | 4.0 |
A TRIR of 4.0 means the equivalent of 4 recordable cases per 100 full-time workers per year. Lower is better.
A few practical points that trip people up:
- Hours worked is actual hours paid for work — exclude vacation, holiday, and sick leave. Estimating from headcount overstates the denominator and understates your rate.
- One incident, one case. A single event injuring two workers produces two recordable cases.
- The metric is lagging. TRIR tells you what already happened. It responds slowly to interventions and, in low-incident operations, can stay flat through luck rather than genuine improvement — a limitation we cover in our incident trend analysis guide.
DART: Days Away, Restricted, or Transferred
DART (Days Away, Restricted, or Transferred) is the rate of recordable cases that resulted in time away from work, restricted duty, or a job transfer, per 100 full-time workers. It isolates the more serious cases — the ones that disrupt the workforce — from the full recordable count.
The formula mirrors TRIR but counts only the qualifying subset:
DART = (Number of DART cases × 200,000) ÷ Total hours worked
A DART case is any recordable injury or illness that produced at least one day away from work, restricted activity, or a transfer to another job. Because every DART case is also a recordable case, DART is always equal to or lower than TRIR (source: EHS Insight, DART Rate, accessed 2026).
Worked example. Using the same facility — 5 recordable cases and 250,000 hours — assume 3 of those cases involved days away or restricted duty.
| Input | Value |
|---|---|
| DART cases | 3 |
| Hours worked | 250,000 |
| Calculation | (3 × 200,000) ÷ 250,000 |
| DART | 2.4 |
The same operation posts a TRIR of 4.0 and a DART of 2.4. The gap between them tells you how much of your recordable burden is serious versus minor.
Why both metrics matter together. TRIR alone can mislead. Two facilities with identical TRIRs can have very different DART rates — one accumulating minor medical-treatment cases, the other producing fewer but more disabling injuries. Reading TRIR and DART side by side shows you not just how often people are getting hurt, but how badly. A widening DART-to-TRIR ratio is a signal that your serious-injury exposure is rising even if total counts look stable.
EMR: The Experience Modification Rate
EMR (Experience Modification Rate, also called e-mod or x-mod) is an insurance metric that compares your workers' compensation claims history to the expected losses for companies of your size and class. Unlike TRIR and DART, it is calculated by an insurance rating bureau — in most U.S. states, the National Council on Compensation Insurance (NCCI) — not by you.
The core relationship is a ratio:
EMR = Actual losses ÷ Expected losses
An EMR of 1.0 is the industry baseline — your claims history matches what's expected for your class and size. Below 1.0 means you outperform the baseline and pay a reduced premium; above 1.0 means you pay a surcharge. An EMR of 0.80 means roughly 20% below baseline premium; an EMR of 1.30 means roughly 30% above (source: Safety Evolution, EMR Rating, accessed 2026).
How the calculation differs from TRIR and DART:
| Feature | TRIR / DART | EMR |
|---|---|---|
| Who calculates it | Your organization | NCCI or a state rating bureau |
| What it measures | Injury frequency / severity | Claims cost vs. expected cost |
| Time window | One calendar year | Three years, excluding the most recent |
| Baseline | Lower is better; no fixed center | 1.0 is average |
| Used for | OSHA reporting, prequalification | Workers' comp premiums, bid eligibility |
Frequency beats severity. The NCCI formula splits each claim into a primary portion (capped — around $18,500 in most NCCI states, adjusted periodically) and an excess portion above the cap. Primary losses are weighted far more heavily, which means claim frequency drives your EMR more than the size of any single claim (source: SmartQHSE, EMR Calculator, accessed 2026). Ten small claims hurt your mod more than one large one of equal total cost. This is why near-miss and minor-injury programs that reduce claim frequency move the EMR needle.
The three-year lag. EMR uses three years of data and excludes the most recent policy year, so a bad year follows you for years and improvements take time to register. The corollary: the corrective actions you implement today show up in your mod two to three years out, not next quarter.
Cut claim frequency, not just claim cost
Because EMR is driven by how often claims occur, reducing recordable frequency is the highest-leverage move you can make. WhyTrace Plus links each incident to root cause analysis and tracked corrective actions, so the same failure does not generate a second claim. Explore WhyTrace Plus →
Benchmarking Against National Data
Benchmarking puts your rates in context — a TRIR of 4.0 is excellent in some industries and poor in others. The Bureau of Labor Statistics publishes national incidence rates annually through its Survey of Occupational Injuries and Illnesses (SOII), giving you a reference point for private industry as a whole.
As of the 2024 data released January 22, 2026, the national private-industry figures were:
| Metric | 2024 rate | Prior year |
|---|---|---|
| TRIR (total recordable) | 2.3 | — |
| DART | 1.4 | 1.5 (2023) |
| DAFW (days away from work) | 0.8 | 0.9 (2023) |
| DJTR (restricted/transfer) | 0.5 | 0.6 (2023) |
Both the TRIR of 2.3 and DART of 1.4 were the lowest on record going back to 2003 (source: BLS Table 1, 2024 industry rates, accessed 2026; BLS Table 2, DART results, accessed 2026).
How to use these numbers honestly:
- Compare within your industry, not against the all-industry average. Construction, warehousing, and manufacturing carry higher baseline rates than office-based sectors. BLS publishes rates by NAICS code; use the one that matches your operation.
- A single year is noise, not a trend. Especially for small employers, one or two incidents can swing TRIR dramatically year to year. Track the multi-year direction.
- Benchmarks are a floor, not a goal. Matching the industry average means you are average. Programs that drive sustained improvement focus on leading indicators — near-miss reporting, hazard closure, corrective-action completion — that move lagging metrics like TRIR over time.
For more on which forward-looking indicators predict injury rates, see our overview of 2026 safety management trends.
Common Calculation Mistakes That Distort Your Rates
The metrics are simple arithmetic, but the inputs are where errors creep in — and a wrong rate on a prequalification form or OSHA submission carries real consequences. These are the mistakes that show up most often.
- Counting first-aid cases as recordable. Inflates TRIR. First aid (per OSHA 29 CFR 1904.7) is explicitly not recordable. Maintain a clear recordability decision log.
- Using scheduled hours instead of actual hours worked. Including paid leave in the denominator deflates your rate. Pull actual hours from payroll.
- Forgetting that DART is a subset of TRIR. If your reported DART exceeds your TRIR, the calculation is wrong by definition.
- Mixing calendar periods. TRIR/DART are annual; EMR uses a three-year window excluding the most recent year. Don't compare a current-quarter TRIR to a lagging EMR and expect them to align.
- Treating EMR as something you calculate. You verify it on the NCCI worksheet your insurer provides — you don't compute it yourself. What you can do is audit the loss runs feeding it for closed claims still listed as open, which can keep your mod artificially high.
- Ignoring the severity signal in the TRIR–DART gap. A low TRIR with a high DART-to-TRIR ratio means your few injuries are serious ones — a different problem than high-frequency, low-severity exposure, and it demands a different response.
For a deeper look at how injury rates translate into financial impact, see our breakdown of incident cost and the business case for prevention.
Frequently Asked Questions
Q. What is a good TRIR or DART rate?
There is no universal target — "good" depends on your industry. As of the 2024 BLS data (released January 2026), the national private-industry TRIR was 2.3 and DART was 1.4. Beating your industry-specific BLS rate, and trending downward over multiple years, is a more meaningful standard than hitting a single round number.
Q. Why is my EMR above 1.0 even though my TRIR is low?
Because EMR weights claim frequency and cost over a rolling three-year window, not single-year recordable counts. A few costly or frequent workers' comp claims in the prior three years can push your mod above 1.0 even while your current TRIR looks strong. The two metrics measure different things over different periods.
Q. Do TRIR and DART use the same 200,000-hour multiplier?
Yes. Both use the 200,000-hour base, which represents 100 full-time employees working a year. The only difference is the numerator: TRIR counts all recordable cases, while DART counts only cases with days away, restricted duty, or job transfer. That is why DART is always less than or equal to TRIR.
Q. How often is EMR recalculated?
Annually, on your workers' compensation policy renewal date. The rating bureau (NCCI in most states) uses three years of loss and payroll data, excluding the most recent policy year, so improvements take two to three years to fully register in your mod.
Q. Can I lower my EMR quickly?
Not quickly — the three-year window means changes lag. The fastest legitimate levers are reducing claim frequency (the most heavily weighted factor), implementing return-to-work programs to limit lost-time claims, and auditing your loss runs to correct claims that were resolved but still reserved as open.
Key Takeaways
- TRIR = (recordable cases × 200,000) ÷ hours worked measures total recordable injury frequency per 100 full-time workers. The 2024 national private-industry average was 2.3.
- DART = (DART cases × 200,000) ÷ hours worked counts only the serious subset — days away, restricted, or transferred — and is always equal to or lower than TRIR. The 2024 national average was 1.4.
- EMR = actual losses ÷ expected losses, centered on a 1.0 baseline, is calculated by NCCI over a three-year window and drives your workers' comp premium. Claim frequency matters more than claim size.
- The most common errors are counting first-aid cases as recordable, using scheduled rather than actual hours, and treating EMR as a number you compute rather than verify.
- Benchmark against your industry-specific BLS rate and the multi-year trend, not a single year or the all-industry average — and use leading indicators to drive the lagging metrics down.
Related Resources
| Resource | Description | Best For |
|---|---|---|
| Incident Trend Analysis: Discovering Seasonal and Shift Patterns | Methods for finding actionable patterns inside your existing safety data | Turning recorded TRIR/DART data into prevention insight |
| The True Cost of Workplace Incidents: Building the Business Case | How injury frequency and EMR translate into direct and indirect cost | Justifying safety investment to finance and leadership |
| Safety Management Trends 2026 | The shift from lagging to leading indicators in EHS measurement | EHS managers modernizing how they measure performance |
For frontline reporting that feeds these metrics, see near-miss and hiyari-hatto reporting workflows (AnzenPost Plus). To strengthen the root cause analysis behind your corrective actions, explore root cause analysis and quality improvement methods (GenbaCompass).
Sources:
- Calculating OSHA Incident Rates: TRIR, DART, LTIFR, and LTIIR | Creative Safety Supply
- DART Rate: Calculating DART the Easy Way | EHS Insight
- TABLE 1. Incidence rates of nonfatal occupational injuries and illnesses by industry, 2024 | BLS
- TABLE 2. DART, DAFW, and DJTR rates, private industry, 2023-24 | BLS
- Experience Modification Rate: EMR Explained | Safety Evolution
- EMR Calculator 2026: Experience Modification Rate (NCCI) | SmartQHSE
- ABCs of Experience Rating | NCCI