The standard incident rate formula is: (Number of recordable injuries and illnesses × 200,000) ÷ Total hours worked by all employees. The result tells you how many incidents occurred per 100 full-time workers over a year, giving you a single number to track safety performance and compare against industry benchmarks.
What the Formula Means
The 200,000 in the formula isn’t arbitrary. It represents the total hours 100 employees would work in a year (40 hours per week × 50 weeks × 100 workers). By standardizing every calculation against this baseline, companies of vastly different sizes can compare their rates directly. A 20-person shop and a 5,000-person factory both end up with a number scaled to the same reference point.
This rate is often called the Total Recordable Incident Rate, or TRIR. OSHA uses it, the Bureau of Labor Statistics publishes industry averages based on it, and most companies report it internally and to clients or regulators.
What Counts as a Recordable Incident
The numerator in your formula is the total number of OSHA-recordable injuries and illnesses during the period you’re measuring. Not every workplace injury qualifies. OSHA considers an injury or illness recordable if it results in any of the following:
- Death
- Days away from work
- Restricted work or job transfer
- Medical treatment beyond first aid
- Loss of consciousness
A few conditions are always recordable at the time of diagnosis regardless of whether they lead to missed work or treatment: cancer, chronic irreversible disease, fractured or cracked bones, and punctured eardrums. If a worker breaks a finger but the doctor says they can keep working without restrictions, that fracture still goes in your count.
Minor injuries treated with basic first aid, like a bandage or an ice pack, do not count. The dividing line is whether the treatment goes beyond first aid, such as prescription medication, stitches, or physical therapy.
How to Count Total Hours Worked
The denominator is every hour actually worked by every employee during the period. This includes hours from salaried workers, hourly workers, part-time staff, seasonal employees, and temporary workers you supervise, even if they’re supplied by a staffing agency.
Do not include vacation time, sick leave, holidays, or any other non-work time, even if employees were paid for those hours. You need actual hours on the job. If your payroll system only tracks hours paid rather than hours worked, or if you have salaried employees without timesheets, OSHA says you must estimate their actual hours worked. For a salaried employee working a standard schedule, 2,000 hours per year (40 hours × 50 weeks) is a reasonable estimate.
A Step-by-Step Example
Say your company had 4 recordable injuries over the past year. You employ 75 people, and your payroll records show a combined 148,000 hours actually worked during that period.
Plug the numbers in: (4 × 200,000) ÷ 148,000 = 5.4. Your TRIR is 5.4, meaning for every 100 full-time workers, you’d expect about 5.4 recordable incidents per year at that rate. For context, the 2024 national average for manufacturing was 2.7 and for construction was 2.2, so a rate of 5.4 would signal room for improvement in either of those industries.
DART Rate and Lost Time Rate
TRIR captures all recordable incidents, but two narrower rates focus on more serious cases. Both use the same formula structure with 200,000 as the multiplier; they just change which incidents go in the numerator.
The DART rate (Days Away, Restricted, or Transferred) counts only incidents that resulted in an employee missing work, being placed on restricted duties, or being transferred to a different job. You exclude cases that only involved medical treatment or loss of consciousness without any lost or restricted time. The 2024 DART rate for construction was 1.3 per 100 workers, while manufacturing came in at 1.7.
The Lost Time Incident Rate, or LTIR, is even narrower. It counts only incidents where an employee missed at least one full day of work because of the injury or illness. The formula is: (Number of lost time cases × 200,000) ÷ Total hours worked. Cases where a worker was given light duty but never missed a day don’t count here.
These three rates together give a layered picture. TRIR shows the overall frequency of recordable events. DART tells you how often those events are serious enough to affect someone’s ability to do their normal job. LTIR isolates the cases that kept someone home entirely.
How to Benchmark Your Rate
A number by itself doesn’t tell you much. You need to compare it against your own industry’s average to know whether your workplace is performing well or poorly. The Bureau of Labor Statistics publishes incidence rates broken down by industry each year. As a reference point from 2024 data: construction averages 2.2 total recordable cases per 100 workers, and manufacturing averages 2.7. Industries with less physical hazard exposure tend to fall well below 1.0.
If your TRIR is lower than your industry average, your safety performance is better than the typical employer in your field. If it’s higher, that gap quantifies how much additional risk your workers face compared to the norm. Many companies also track their own rate over time, year by year, to see whether new safety programs or operational changes are actually reducing incidents. A single year’s rate can fluctuate, especially for smaller companies where one or two incidents swing the number significantly, so a three-year trend is more telling than any single calculation.
Common Mistakes to Avoid
The most frequent error is inflating the denominator with paid hours that weren’t actually worked. Including vacation and holiday pay makes your hours total larger, which artificially lowers your rate. Stick to actual hours on the job.
Another common mistake is leaving out temporary or contract workers you supervise. If you direct their day-to-day work, their hours and their injuries belong in your calculation, not their staffing agency’s.
Finally, be careful with the time period. If you’re calculating a quarterly rate rather than an annual one, the formula stays the same. You use the incidents and hours from that quarter only. The 200,000 multiplier doesn’t change, because its job is to normalize the result to a per-100-worker annual equivalent regardless of the time window you’re measuring.