By Rovaryn Digital · 9 min read
Why Your Injury Rate Needs a Reference Point
The safety report hits your desk in January. Your DART rate is 1.8. Is that good? Acceptable? A number that should prompt an immediate program review? Without a reference point, it is impossible to say — and neither your carrier nor your workers' compensation broker will offer one unprompted.
That is the practical problem injury rate benchmarking solves. When you map your rates against your NAICS peer group — the companies doing the same work, hiring from the same labor pool, facing the same physical exposures — you move from a number in isolation to a position on a spectrum. A DART rate of 1.8 in light manufacturing looks very different from a DART rate of 1.8 in a skilled-nursing facility or a general contractor's field operation.
The same logic applies to return-to-work outcomes. The share of injured workers back on modified duty within 30 days, the average days to first transitional assignment, the percentage of claims that cross from medical-only into lost-time status — each of these carries meaning only when measured against what comparable employers achieve. Benchmarks give you the reference; your own data gives you the gap to close.
This article explains where the public benchmark data lives, how to calculate the rates your NAICS peers are measured on, and how to read your position honestly — including the limits of the comparison.
The Two Public Data Sources Worth Using
BLS Survey of Occupational Injuries and Illnesses (SOII)
The Bureau of Labor Statistics publishes annual injury and illness counts and rates for private-sector employers through the Survey of Occupational Injuries and Illnesses. Rates are published at the 3-digit and 4-digit NAICS level, which means you can find a rate for, say, food manufacturing or general freight trucking rather than only "all manufacturing" or "transportation."
The two rates you care about most:
DART rate (Days Away, Restricted, or Transferred rate). Cases per 100 full-time equivalent workers where the injury resulted in at least one day away from work, one day of job transfer, or one day of work restriction. This is your primary EMR-adjacent operational metric, because DART cases are the ones most likely to generate indemnity costs and modifiable-duty demand.
TRC rate (Total Recordable Case rate). All OSHA-recordable injuries and illnesses per 100 FTE, regardless of severity. Useful for program-level trend analysis; less directly tied to lost-time economics than DART.
The 2024 BLS release reported a total recordable case rate of 2.3 per 100 FTE across all private-sector industries, the lowest in the series, based on 2024 data (BLS, 2026). The 2023 data showed 946,500 days-away cases at a rate of 0.9 per 100 FTE (BLS TED, 2025). These all-industry figures are floors for comparison — your sector rate may be materially higher.
Sector-specific examples from the verified data: construction recorded 2.3 injuries per 100 FTE in 2023 (BLS via Work Comp Professionals, 2024), and healthcare and social assistance saw physical injury cases rise to 471,600 cases in 2023 (BLS via Work Comp Professionals, 2024). Transportation and warehousing recorded 265,700 total cases in 2023 (BLS via Work Comp Professionals, 2024).
To find your specific subsector rate, go to bls.gov/iif and download the SOII industry-specific tables for the most recent survey year. Match your 4-digit NAICS code; if your code is not published at the 4-digit level, use the 3-digit parent. Confirm the survey year of the table — BLS publishes a year behind, so 2024 actuals will appear in late 2025 or early 2026 releases.
OSHA Injury Tracking Application (ITA)
OSHA's ITA collects establishment-level Form 300A summary data from employers required to electronically submit. The resulting dataset — available at osha.gov/establishment-specific-injury-and-illness-data — lets you look at individual establishment records filtered by NAICS code, state, and establishment size.
Where SOII gives you a rate derived from a statistical survey sample, ITA gives you actual submitted records. This means you can find establishments in your 6-digit NAICS, your state, and your size band, and see what their 300A-reported rates look like. That is a more granular peer comparison than SOII alone.
Practical limitations: ITA data reflects what employers self-reported on their 300A summaries. Under-recording of minor injuries is a known issue across industries, so treat ITA rates as directional rather than definitive. Use ITA alongside SOII, not in place of it.
Calculating Your Own Rates
Before you can benchmark, you need to compute your rates on the same basis the BLS uses — otherwise you are comparing unlike numbers.
FTE denominator. BLS uses 200,000 hours as the equivalent of 100 full-time workers for one year (100 workers × 2,000 hours). Every rate calculation divides by your total hours worked and multiplies by 200,000.
DART rate formula:
DART Rate = (DART cases × 200,000) ÷ Total hours worked
Worked example: An employer with 180 workers logs 360,000 hours in a calendar year and records 7 DART cases. DART Rate = (7 × 200,000) ÷ 360,000 = 3.9. That is a rate to compare against the sector SOII table.
TRC rate formula:
TRC Rate = (Total recordable cases × 200,000) ÷ Total hours worked
Pull hours from payroll, not from headcount estimates. Include overtime. Exclude hours for temporary workers supplied by a staffing agency if they are recorded on the agency's 300 log.
Days Away rate (DAWR) — cases with at least one day away from work only:
DAWR = (Cases with days away × 200,000) ÷ Total hours worked
Your DART rate will always be at or above your DAWR. The spread between them is your restricted-duty and job-transfer volume — the population of workers your transitional-duty program is already serving, or could be.
Reading Your Position Against the Benchmark
Once you have your rate and your sector's SOII rate for the same metric year, the comparison is straightforward arithmetic. What requires judgment is interpreting it correctly.
Below the sector rate. Your program, selection, or physical environment is outperforming peer averages on frequency. That does not mean the program is optimized — severity and RTW speed may still lag. A medical-only-to-lost-time conversion rate above sector norms alongside a low TRC rate often means injuries are being recorded but not managed aggressively into modified duty before the indemnity clock starts.
At or near the sector rate. You are running with the pack. Whether that is acceptable depends on whether your sector is itself high-risk and whether your EMR is trending toward or away from 1.0. For context, an EMR of 1.0 is the industry average by definition; an EMR of 1.3 on a $10,000 base premium becomes $13,000 (Berry Insurance, 2024). Moving meaningfully below the sector rate has direct premium consequences. See the experience modification rate explained for a full breakdown of the EMR mechanics.
Above the sector rate. Frequency is a bigger EMR driver than severity. Five $10,000 claims raise EMR more than one $50,000 claim (PolicyBenchmark, 2026), which means a rate above your peer group is likely costing you more in premium load than the raw claim dollars suggest.
One structural note: the EMR experience window covers three years excluding the current policy year (Higginbotham, 2026). A rate spike this year will not leave the window for roughly three years. That lag makes early-stage benchmarking and program correction more urgent, not less.
Adding RTW Outcome Benchmarks
Injury rate data tells you how often workers get hurt. RTW outcome data tells you what happens next — and that is where transitional-duty program quality actually shows up.
The public benchmark on RTW timing: approximately 50% of injured workers return within 30 days, and approximately 75% return within three months (WCRI, 2018). Those figures are aggregate across all industries and employer sizes.
Employer size has a pronounced effect on RTW outcomes. The share of workers not returning to the pre-injury employer is 21% at employers with 1–50 workers, compared to 10% at employers with 251–1,000 workers and 7% at employers with 1,000 or more (WCRI, 2018). Smaller operations have structurally fewer transitional assignments available, which makes a documented transitional-duty inventory — maintained in advance, not assembled after the injury — a meaningful competitive advantage for smaller employers.
Duration off work compounds the problem non-linearly. RTW likelihood drops to approximately 50% after 45 days off work (RACP/AFOEM, 2010). That benchmark defines the window your program needs to operate inside.
Sector-level RTW benchmarks broken out by NAICS are not available in the public data at the specificity of SOII injury rates. If your carrier or TPA provides loss-run data with RTW timing, compare your average days-to-modified-duty against the WCRI aggregate figures and against your own trend year over year. Your own trend is often the more actionable benchmark.
For a structured approach to building out the transitional-duty inventory that feeds RTW performance, the return-to-work case management guide covers the program mechanics. Industry-specific transitional task frameworks are available for manufacturing and construction operations.
The Limits of Benchmarking
Benchmarks are diagnostic tools, not scorecards. Three limitations matter in practice.
Survey lag. SOII data runs one to two years behind the current calendar year. Rates published today reflect workplace conditions from the prior year. If your operation changed significantly — new equipment, a workforce reduction, a new product line — the peer rate may not yet reflect those changes in the sector either.
Mix and size effects. A 4-digit NAICS rate averages across operations of all sizes and geographies. A small specialty contractor in a high-cost state will not benchmark cleanly against a large regional general contractor in the same NAICS code. Use the rate as orientation, not as a precise target.
Recording practice variation. Both SOII and ITA depend on consistent OSHA recordkeeping. If your recording practice is more rigorous than the sector average — you record borderline cases where others might not — your rate will appear elevated against peers even if your actual injury experience is comparable. Audit your recordkeeping practice before drawing conclusions from an unfavorable benchmark.
Turning Benchmark Data Into a Budget Argument
An unfavorable DART rate relative to NAICS peers, combined with the WCRI RTW timing data and the EMR mechanics, gives you the structure of a program investment argument: frequency above sector norms multiplies premium costs, and lost-time claims that outlast the 30-day window drive both EMR elevation and the severity curve.
The RTW Program ROI & Budget Case Workbook provides a structured template for translating your benchmark position — your rate versus the sector rate, your current EMR, your average claim cost — into a projected ROI on transitional-duty program investment. It is a practical complement to the benchmarking exercise this article describes.
Stay Current on RTW Program Metrics
BLS releases updated SOII tables annually, typically in the fall. OSHA updates the ITA dataset periodically. Benchmarking is not a one-time exercise — your sector rate moves, your own rate moves, and the gap between them is what the program is managing against.
If you want a digest of updated benchmark data and RTW program guidance as it becomes available, the Transitional Duty Manager newsletter covers new BLS and OSHA releases alongside practical program content. Subscribe below to get the next update when SOII rates are refreshed.
Get the next RTW guide in your inbox
Practical guides and WA Stay-at-Work updates — no spam.
Automate the full RTW workflow
Transitional Duty Manager replaces manual RTW documentation with O*NET duty matching, WA SAW reimbursement packet export, and an immutable audit trail.
See how it works
