Sports/recreation, delivery, nursing care, and correctional facilities post the highest workers-comp claim proxy rates; the injury mix is dominated by overexertion and contact injuries

Summary

No free government source publishes actual workers-compensation claim rates by industry (NCCI's class-code claim-frequency data is proprietary). The standard public proxy is BLS's Survey of Occupational Injuries and Illnesses (SOII), which tracks employer-reported recordable injuries/illnesses (TRC) and the more serious subset involving days away, job transfer, or restriction (DART) — DART cases are the ones most likely to trigger an indemnity workers-comp claim. By that measure, the highest-rate detailed industries in 2024 are sports teams & clubs (16.5 TRC per 100 full-time workers, 7x the 2.3 national average), veterinary services (11.1), skiing facilities (10.0), local messengers/delivery (8.5), state-government nursing/residential care (8.3), couriers & express delivery (7.9), and correctional institutions (7.5) — a mix dominated by high-exertion recreation/animal-handling work, last-mile delivery, and direct patient/inmate care. On the DART measure, delivery/courier services and air transport move up alongside recreation and nursing care. Nationally, rates have been falling for over a decade (TRC 3.4 in 2012 to 2.3 in 2024), consistent with NCCI's own report of a ~6% drop in workers-comp lost-time claim frequency in accident year 2024. Within that decline, the type of injury behind claims has shifted only modestly and is genuinely hard to trend precisely: overexertion/repetitive-motion/bodily-condition injuries remain the single largest category of lost-time cases (32% of DART cases, 2023-24), but contact incidents (29%) and falls/slips/trips (24%) are now nearly as large, and BLS revised its injury-classification system in this release (OIICS version 3), creating a formal break in the series that limits exact multi-year comparison of injury-type shares.

Which industries have the highest claim-proxy rates

BLS's SOII publishes two headline severity measures per 100 full-time-equivalent (FTE) workers: Total Recordable Cases (TRC — every recordable injury/illness) and DART (the subset involving days away from work, job transfer, or restriction — the injuries serious enough to typically generate a workers-comp indemnity claim, not just a first-aid/medical-only record). BLS's own "highest rate" tables (Table SNR01 and SNR02, restricted to detailed industries with at least 500 recordable cases) for 2024 show:

On the DART measure — arguably the tighter claim proxy, since these cases involve real lost time — recreation, delivery, and direct-care industries again lead: sports teams/clubs (11.0), skiing facilities (7.3), local messengers/delivery (6.9), couriers/express delivery (6.5), scheduled passenger air transport (6.4), and state-government nursing/residential care (6.1), against a 1.5 national DART average.

At the broader supersector level, the 2024 SOII news release states health care and social assistance ran a TRC rate of 3.4 (down from 3.6 in 2023) — still well above the 2.3 all-industry average — while no sector's rate rose in 2024; declines were recorded in information, health care and social assistance, retail trade, manufacturing, and real estate/rental and leasing.

Has the injury-type mix behind those claims been changing

For 2023-24 combined, private-industry DART cases break out by event/exposure as: overexertion, repetitive motion, and bodily conditions — 946,290 cases (44.7 per 10,000 FTE, 32% of DART); contact incidents — 860,050 cases (40.6, 29%); falls, slips, trips — 721,720 cases (34.1, 24%); exposure to harmful substances/environments — 224,450 cases (10.6, 7%; notably 88% of these required at least one day away from work, the highest such share of any category); transportation incidents — 121,330 cases (5.7, 4%); and violent acts — 78,340 cases (3.7, 3%).

Overexertion/repetitive-motion injuries (sprains, strains, tears, and other musculoskeletal disorders) have been the largest single category for well over a decade — BLS's 2015 release put sprains/strains/tears at 37% of days-away-from-work cases — so the identity of the leading cause has not changed. What has narrowed is the gap: contact incidents and falls/slips/trips are now close behind rather than clearly secondary, though a clean numeric trend line across the two eras is not available, because BLS implemented a revised injury classification system (OIICS version 3) starting with this 2023-24 release, which the agency itself flags as creating a formal break in the nature/event-of-injury series. This release also added, for the first time, a specific "coronavirus — novel" nature-of-injury code (annualized DAFW rate 5.6 per 10,000 FTE in 2023-24, concentrated in healthcare support and healthcare practitioner occupations at 32.4 and 26.7 respectively) — a genuinely new category rather than a shift within the old ones.

Corroborating the overall downward claim trend independently: NCCI (the workers-comp insurance industry's own rating bureau) reported that accident-year 2024 claim frequency is expected to run about 6% below 2023, and that lost-time claim frequency specifically fell about 5% — consistent in direction and rough magnitude with BLS's TRC/DART declines over the same period, even though NCCI's underlying claim-frequency-by-class-code data is not itself publicly queryable.

Data and methodology notes

• No free public source of true workers-comp claim rates by industry exists. NCCI publishes claim frequency (claims per $100 of payroll) by class code through a subscriber-only Industry Trend Dashboard; state workers-comp bureaus publish loss-cost/rate filings, not national industry claim-rate tables. BLS SOII TRC/DART incidence rates are the standard, widely-used public substitute, because OSHA-recordable injuries serious enough to cause lost time are the population from which most indemnity claims are drawn.
• "Highest rate" industry lists are restricted by BLS to detailed NAICS industries with at least 500 total recordable cases at the most detailed level of publication, so very small industries with high rates but low case counts are excluded by BLS's own table design, not by this analysis.
• The AskAmerica warehouse (econ/census schemas) does not carry SOII injury/illness data — it holds QCEW employment/wages and ACS industry employment, neither of which includes injury or claim counts — so this analysis is built entirely from BLS's primary published SOII tables (fetched via Wayback Machine snapshots after BLS's live pages returned HTTP 403 to automated fetches; the archived captures are the identical BLS-published files, dated January-August 2026).
• Nature/event-of-injury shares before vs. after 2023 are not strictly comparable due to the OIICS version 3 classification revision BLS implemented for the 2023-24 release; qualitative continuity (overexertion/MSD injuries remaining the largest category) is supported by BLS's own historical narrative releases, but exact share-point changes across the revision boundary should not be read as precise.