Out of more than 500 jobs in the United States, the two with the highest share of deaths from radiation-linked cancers belong to flight attendants and pilots. Not nuclear technologists. Not people who work around X-ray machines. The folks who hand you a ginger ale.

That's the takeaway from a study published August 17 in JAMA Internal Medicine. It was led by Vishal Patel, a clinical fellow in surgery at Brigham and Women's Hospital, with senior author Anupam Jena of Harvard Medical School. Their conclusion: it may be time for the U.S. to protect aircrew the way it protects other workers who deal with radiation.

Why anyone worried about this in the first place

Aircrew soak up more ionizing radiation on the job than almost any other American workforce. It comes from space. Way up at cruising altitude, around 33,000 feet, particles from the sun and the wider galaxy slam into the atmosphere and shatter into a shower of neutrons and gamma rays. Roughly 40 to 65 percent of the dose crews get comes from the heavy-hitting kind of radiation, mostly neutrons.

Earlier research had already linked flying for a living to higher rates of several cancers. Melanoma showed up most often, along with breast cancer, brain and nervous system cancers, and blood cancers. But nobody had proven that cosmic radiation was the culprit, partly because it's hard to untangle radiation from everything else that comes with the job.

What the new study did

The researchers used U.S. death certificates from 2020 through 2024, which have only recently been linked to what people did for a living. That's 12,710,517 deaths across 503 occupations, including 14,190 pilots and 7,170 flight attendants.

They then adjusted the numbers for the usual suspects that mess with mortality comparisons: age at death, sex, race, ethnicity, education, and marital status. The cancers they counted as radiation-related included breast, brain and nervous system, multiple myeloma, most leukemias, lymphoma, thyroid, prostate, melanoma, and other skin cancers.

The numbers

Both pilots and flight attendants died more often than expected from breast cancer, brain and nervous system cancers, prostate cancer, and melanoma. Pilots also had more leukemia deaths.

Add it up and about 6.9 percent of flight attendant deaths and 6.7 percent of pilot deaths came from radiation-related cancers. First and second place out of 503 jobs.

Nuclear technologists, who spend their careers around the kind of radiation that comes with warning signs and lead aprons, came in twelfth.

Two details make the finding harder to wave away. First, the pattern showed up only in people who actually fly. Aircraft mechanics and assemblers, who work on the same planes but stay on the ground, showed nothing unusual. Second, for cancers not linked to radiation, aircrew landed right around the middle of the pack. Whatever is going on, it seems to be happening in the air.

The rules haven't caught up

Crews pick up an estimated 3 to 6 millisieverts of cosmic radiation per year, depending on where they fly. For comparison, the CDC estimates a passenger taking 10 cross-country round trips a year gets about 0.4 millisieverts. Crews get roughly ten times more.

The International Commission on Radiological Protection suggests capping crew exposure at 20 millisieverts a year, and at 1 millisievert over an entire pregnancy.

Here's the odd part. The FAA officially recognizes pilots and flight attendants as workers exposed to ionizing radiation. And then it stops there. Unlike other radiation-exposed workers in the U.S., and unlike aircrew in plenty of other countries, American crews face no federal dose limits and no monitoring requirements. They are officially at risk and unofficially on their own.

Does flying at night help? Sort of.

A lot of people assume daytime flights mean more radiation. For cosmic radiation, that's mostly a myth. What drives the dose is altitude (it roughly doubles for every 5,000 feet you climb), latitude (lowest near the equator, highest near the poles, because Earth's magnetic field shields the middle better), solar activity, and how long you're up there. The clock has nothing to do with it. A red-eye over the Arctic and a noon flight over the Arctic at the same altitude get similar doses.

What does change after dark is ultraviolet light. Cabin and cockpit windows let UVA through, and up high the atmosphere is thinner and clouds and snow bounce sunlight right back at you. UV is a proven cause of melanoma and other skin cancers, and it's a leading explanation for all that extra melanoma in crews. Layovers in sunny places don't help either.

So the honest answer is: night flying won't lower your cosmic dose, but it does erase your UV dose.

Two things actually do cut cosmic exposure, and both are operational choices: fly lower, and skip polar routes. The FAA also issues solar radiation advisories during big solar storms so airlines can adjust altitude or routing. For UV, the fixes are refreshingly low-tech. UV-filtering window film, long sleeves, and broad-spectrum sunscreen on daytime legs.

The fine print

The authors are upfront about what they couldn't do. They couldn't measure how many people were in each job overall, and they couldn't measure anyone's actual radiation dose. Occupations on death certificates can be wrong. And flying comes with other hazards they didn't measure, including wrecked sleep from night flights and time zone hopping, plus engine bleed air.

There's also contrary evidence. Some large studies, including one following German aircrew, found no clear link between how much radiation crews got and how much cancer they developed. That matters, because a real cause usually shows more effect with more exposure.

Still, the fact that the signal appears in flying crews and not in the mechanics standing next to the same aircraft is hard to explain away. It isn't proof. It is a good argument for finally measuring the dose these workers are taking.

REFERENCES
Patel, V., et al. (2026). Cancer-related mortality among US pilots and flight attendants. JAMA Internal Medicine. https://doi.org/10.1001/jamainternmed.2026.3635