And Then You're Dead

The Autopsy / Case File · Workplace & Occupational

The Deadliest Job

A logger's fatal-injury rate runs 33 times the U.S. average. A commercial fisherman's runs 27 times. Neither job generates a fraction of the news coverage of far statistically safer occupations.

The Bureau of Labor Statistics tracks every fatal workplace injury in the country and publishes the fatality rate — deaths per 100,000 full-time-equivalent workers — by occupation every year. The 2024 numbers put logging, commercial fishing and hunting, and roofing at the top of the list, by a wide margin, and none of the three looks anything like the jobs most safety coverage focuses on.

And Then You're Dead · September 2026 · 9 min read

Join the discussion →
A roofer kneels on a steep residential roof, nail gun in hand, installing asphalt shingles
A roofer repairs tornado damage in Norman, Oklahoma. Roofing carries the third-highest fatal-injury rate of any U.S. occupation, at nearly 15 times the national average.

Every year, the Bureau of Labor Statistics publishes the Census of Fatal Occupational Injuries, a count that isn't a survey or an estimate — it's built from death certificates, workers' compensation records, OSHA reports, news accounts, and other administrative sources cross-checked against each other, and it aims to capture every work-related fatality in the country. The 2024 release put total fatal work injuries at 5,070, a 4.0% decline from 2023's 5,283 — a genuine improvement. But the topline number obscures how unevenly that risk is distributed. The overall U.S. fatal work injury rate in 2024 was 3.3 per 100,000 full-time-equivalent workers. Three occupations blew past that baseline by an order of magnitude or more.

Logging: 110.4 Per 100,000

Logging workers led every occupation tracked by BLS in 2024, at a fatal-injury rate of 110.4 per 100,000 full-time-equivalent workers — a rate roughly 33 times the national average. The mechanism is straightforward and largely unchanged across decades of surveillance: felling trees, especially large or already-damaged ones, is inherently unpredictable physics. A tree can split, kick back, or fall in a direction the sawyer didn't plan for, and the same worksite hazards recur case after case in federal fatality investigations — being struck by a falling tree or limb, being pinned by rolling logs on steep terrain, and being crushed by heavy equipment operating in tight, uneven forest conditions with limited sightlines.

0
fatal-injury rate per 100,000 FTE logging workers in 2024 — the highest of any tracked U.S. occupation, roughly 33 times the 3.3-per-100,000 national average
0
fatal-injury rate per 100,000 FTE fishing and hunting workers in 2024 — roughly 27 times the national average, the second-highest of any occupation

Fishing and Hunting: 88.8 Per 100,000

Commercial fishing carries a different profile of hazard, but a comparably brutal rate. Vessel disasters — capsizing, flooding, collision — remain the single largest driver of fatalities in the sector, frequently in remote, cold waters like the Bering Sea and North Atlantic where rescue response times run to hours rather than minutes, if a rescue is possible at all. Falls overboard, entanglement in gear and winches, and onboard equipment failures round out the rest. The industry's structure compounds the physical hazard: many commercial fishing operations are small, owner-operated vessels that fall outside some of the safety-equipment and inspection requirements that apply to larger commercial marine operations, and a bad weather day at sea offers far fewer safe options than a bad weather day on land.

Roofing: 48.7 Per 100,000

Roofers ranked third at 48.7 fatal injuries per 100,000 full-time-equivalent workers in 2024 — 104 deaths in the occupation that year, at a rate nearly 15 times the national average. The mechanism here is almost entirely one category: falls, from the roof edge, through a fragile or unsupported roof surface, or off a ladder while carrying material up or down. Roofing work is seasonal, weather-dependent, frequently performed by small subcontracting crews under time and cost pressure, and — unlike a controlled construction-site fall from scaffolding — often happens on a sloped, unstable surface with no guardrail at all unless the crew specifically rigs one.

A logger's fatal-injury rate is 33 times the national average. A commercial fisherman's is 27 times. A roofer's is nearly 15 times. None of the three occupations generates headlines proportional to that risk.

Just behind the top three, refuse and recyclable material collectors recorded a fatal-injury rate of 37.4 per 100,000 in 2024 — a job whose hazard profile (struck by traffic while working curbside, caught in compactor machinery) gets even less public attention than logging, fishing, or roofing, despite a fatality rate more than eleven times the national average.

A Rate That Fell, Then Climbed Back

Logging's fatality rate isn't a fixed fact of the job — it has moved substantially over the past three decades, in both directions. OSHA's comprehensive logging-operations standard, 29 CFR 1910.266, took full effect in 1995, mandating specific felling procedures, protective chaps, and communication protocols between crew members. In 1997, shortly after the standard took hold, the logging fatality rate stood at 128.7 per 100,000 workers. By 2006, it had fallen to 85.6 per 100,000 — a genuine, substantial improvement that safety researchers credited in large part to the standard along with broader adoption of mechanized harvesting equipment that keeps workers farther from a falling tree.

That decline didn't hold. By 2024, the rate had climbed back to 110.4 per 100,000 — still well below the late-1990s peak, but a meaningful reversal from the mid-2000s low. Occupational safety researchers point to a mix of contributing factors: an aging logging workforce with a shrinking pipeline of younger entrants, continued reliance on smaller independent contractors who may have less capital to invest in the newest mechanized equipment, and the sheer physical unpredictability of manual felling on steep or storm-damaged terrain, which no procedural standard fully eliminates. The lesson embedded in the up-then-down trend line is a blunter one than "the standard worked" or "the standard failed" — logging's floor of danger sits high enough that even a well-implemented regulatory intervention only compresses the risk, and the compression isn't guaranteed to be permanent once other conditions in the industry shift.

Rate Versus Total Count — A Different Kind of Danger

The occupations with the highest per-capita fatality rates are almost never the occupations that produce the largest total number of workplace deaths, because they employ relatively few people. Logging is a small, geographically concentrated workforce; even at 110.4 deaths per 100,000 workers, the absolute logger death count in a given year runs to the dozens nationally, not the hundreds. Transportation and material moving occupations, by contrast, recorded 1,391 total fatalities in 2024 — the most of any occupational group — but spread across a workforce so much larger that the group's fatality rate looks unremarkable by comparison. Both numbers describe something true and different: one measures how dangerous the job is for the person doing it, the other measures where the largest share of the national death toll actually accumulates. A safety story built only around whichever number sounds more alarming in a given year will tell a misleading version of either one.

0
total U.S. fatal work injuries in 2024, down 4.0% from 5,283 in 2023 — a genuine year-over-year improvement, per BLS's Census of Fatal Occupational Injuries
0%
share of all 2024 occupational fatalities classified as transportation incidents — the single most common fatal-event type across every industry combined

Why the Riskiest Jobs Stay Invisible

Part of why logging, fishing, and roofing fatalities don't generate proportional public attention is structural, not accidental. These deaths happen one at a time, in remote or elevated locations with no bystanders to record them, to workers who are frequently non-unionized, seasonally or informally employed, and geographically dispersed across small towns and coastal communities far from major media markets. A mass-casualty industrial accident draws national coverage almost by definition; a roofer falling off a single-family home in a town of 4,000 people does not, even though the roofer was statistically far more likely to die on the job that year than almost anyone else in the American workforce. The BLS's own fatality-rate data exists specifically to correct for that visibility gap — it's the mechanism by which regulators, insurers, and safety researchers know where the disproportionate risk actually concentrates, independent of which deaths happen to make the news.

That data also shapes where safety regulation gets built and enforced. OSHA's construction-focused rules, including the fall-protection standards that most directly address roofing risk, exist because this exact kind of rate data identified falls as the dominant mechanism of death in the trade. Logging and commercial fishing, by contrast, involve enough small, geographically dispersed, and sometimes maritime or tribal-land operations that comprehensive federal safety enforcement remains harder to apply uniformly — a gap the fatality-rate data makes visible even when it doesn't, on its own, close it.

And Then You're Dead

Nobody chooses a career by checking the BLS fatality-rate table first. The people cutting timber on a steep slope, hauling a crab pot in on a pitching deck, or kneeling on a hot asphalt roof with a nail gun are doing ordinary, necessary work that the rest of the country's homes, seafood counters, and paper supply depend on — work that kills at a rate most office employees would find unthinkable if it were attached to their own job title instead of someone else's.

That's it. That's the whole thing.

Comments

Loading…

Comments are anonymous and moderated.

Sources

Photo: "Roofer Working on a Home in Oklahoma," by Patsy Lynch/FEMA, 2010, via Wikimedia Commons (public domain, U.S. government work).