Pilots and the new cancer study: what it shows and what it doesn't

In August 2026, JAMA Internal Medicine published a Harvard study that ranked flight attendants first and pilots second among 503 US occupations for the share of deaths caused by cancers linked to radiation. The result deserves attention, but it measures something narrower than "pilots get more cancer." This piece looks at how the study worked, what it cannot tell you and how it squares with known crew doses.

What the study measured

The paper is a short research letter, Cancer-Related Mortality Among US Pilots and Flight Attendants, by Vishal Patel, Michael Liu and Anupam Jena, published online on August 17, 2026. It is cross-sectional: a snapshot of deaths, not a group of workers followed over time.

The data came from US death certificates for 2020 through 2024 in the National Vital Statistics System, which records each person's usual occupation. According to Harvard Medical School, that covered nearly 13 million deaths across 503 occupations, including about 14,000 pilots and 7,000 flight attendants.

"Radiation-related cancer" had a specific meaning here: breast, central nervous system, multiple myeloma, leukemia (except chronic lymphocytic leukemia), lymphoma, thyroid, prostate, melanoma and non-melanoma skin cancers. Lung cancer was left out on purpose, to limit the effect of smoking.

The main measure was the proportion of each occupation's deaths that came from those cancers. The researchers adjusted for age at death, sex, race, ethnicity, education and marital status, then compared occupations. As checks, they used nuclear technologists as a group expected to show a signal, and aircraft mechanics and assemblers as aviation workers who stay on the ground, according to the ASCO Post summary.

What it found for pilots

Nuclear technologists placed 12th. Aircraft mechanics and assemblers showed no elevation, which suggests the signal is tied to flying rather than to aviation work in general.

By cancer type, the Harvard release reports statistically significant elevations in both groups for breast, central nervous system and prostate cancers and for melanoma. For pilots, the odds ratios were 1.57 for breast cancer, 1.38 for prostate cancer and 1.28 each for central nervous system cancers and melanoma. Leukemia was elevated for pilots only (odds ratio 1.27, 95% CI 1.08 to 1.49). The authors concluded that the findings support considering "occupational radiation protections for U.S. aircrew members commensurate with their level of exposure."

What it cannot show

The first limit is that it measures a share of deaths, not a rate. The study looked only at people who died and asked what fraction died of these cancers. It did not count how many pilots per 100,000 get or die of cancer each year. A proportion can rise because one cause is more common or because other causes are less common. If pilots die less often of other diseases than the average worker, cancer takes a larger slice even at the same underlying rate. An odds ratio of 1.36 compares those slices. It does not mean a pilot has a 36% higher chance of getting cancer.

It also shows an association, not a cause, and it had no individual dose data. In a TIME essay, Patel, Jena and their colleague Christopher Worsham listed the limits themselves. The data cannot show how much radiation any individual received. Occupations are sometimes recorded incorrectly on death certificates. People predisposed to these cancers could be drawn to flying jobs. And crew face other hazards, from disrupted circadian rhythms to in-cabin exposures, that the study could not fully account for.

Other exposures overlap with radiation. CDC/NIOSH says aircrew are more likely than the general population to be diagnosed with melanoma and other skin cancers, and that nobody knows for sure why. UV at cruise altitude is higher than at sea level, glass windshields block a little less than half of UVA while plastic ones block most of it, and crew may also get more sun off duty. Because melanoma and non-melanoma skin cancers are on the study's list, part of the signal could come from UV rather than cosmic radiation. CDC also names circadian disruption as a possible contributor to cancer risk.

The adjustment was demographic. It covered age, sex, race, ethnicity, education and marital status, but not smoking history, sun habits, family history or reproductive history.

Finally, "pilot" is a broad label. A death certificate records a usual occupation, not the type of flying, the routes or the years spent at cruise altitude. People who died between 2020 and 2024 mostly flew their careers years or decades earlier.

How much radiation crew actually get

The Harvard release puts aircrew at an estimated 3 to 6 millisieverts (mSv) a year, depending on routes. The FAA's AC 120-61B cites a 2000 report by the UN Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) that ranked aircrew as the fourth most exposed group of workers, averaging 3 mSv a year. The committee's latest survey puts the worldwide average for 2010 to 2014 at 2.7 mSv (UNSCEAR 2020/2021, Annex D). For scale, a routine chest X-ray is about 0.1 mSv (Mettler et al., Radiology, 2008, two-view exam), so 3 mSv a year is about what 30 chest X-rays deliver.

Single flights vary widely. These CARI-7 estimates come from Table 2 of the circular (ICRP 103 weighting, mean solar activity of January 2000, quiet geomagnetic conditions):

And the reference points:

Typical crew doses sit well below the occupational limits and above the public one. The FAA circular adds that these limits are not thresholds beyond which a dose becomes intolerable; they are upper limits of what is considered acceptable, based on current estimates of risk.

US and EU rules differ

In the US, AC 120-61B is guidance. It supports airline programs that inform crew about in-flight radiation and states recommended limits, but no regulation requires an airline to measure, record or cap crew dose. The Harvard release makes the same point: the FAA recognizes aircrew as occupationally exposed, yet US crew have no federal dose limits or monitoring requirements.

In the EU, Council Directive 2013/59/Euratom, Article 35(3), requires that where crew dose is liable to be above 1 mSv a year, the operator must assess crew exposure, take it into account when building schedules so that highly exposed crew get lower doses, tell crew about the health risks and their individual dose, and apply the directive's pregnancy protections. Where crew dose is liable to exceed 6 mSv a year, the directive's wider occupational exposure requirements apply.

Practical takeaways for pilots

Start by knowing your own number. A range of 3 to 6 mSv covers many different schedules, and where you fall depends on what you fly. US airlines are not required to track it, so the job falls to you. The FAA's CARI program is the reference tool for flight dose estimates.

Know the levers, too. AC 120-61B names time in the air, altitude, latitude and solar activity, and it puts galactic cosmic radiation over the polar regions at about twice the level over the geomagnetic equator at the same altitude. Solar activity works in reverse. An FAA technical report on aerospace radiation explains that the solar wind is at its most intense in the active phase of the cycle, which pushes cosmic radiation in the atmosphere to its lowest levels. NASA and NOAA announced in October 2024 that the Sun had reached the maximum phase of Solar Cycle 25, expected to last another year or so before the decline. As activity falls toward the next minimum, the same route will deliver more dose.

Bidding decisions are personal. CDC/NIOSH suggests that crew who want less exposure try to reduce time on very long flights, high-latitude flights and polar routes, while warning that bidding to cut one exposure can increase another. Those trips also carry pay and schedule trade-offs that only you can weigh. For solar radiation storms, follow your company's procedures and the published guidance.

Skin protection is a separate and simpler step. CDC recommends sunscreen at work, since some UV comes through the windows on some aircraft, especially those with glass windows.

Take health questions to a professional. This article is information, not medical advice. For pregnancy, screening or personal risk, talk to your doctor or aviation medical examiner.

A note for passengers

The Harvard release estimates that someone taking 10 cross-country round trips a year receives about 0.4 mSv. In their TIME essay, the authors put that at roughly a seventh to a fifteenth of what crew members accumulate in a year, and wrote that the study does not say your next flight, or even a lifetime of flying for annual vacations, meaningfully raises your risk of cancer.

What LogAloft does with this

LogAloft calculates the cosmic radiation dose of every flight you log, with a model based on the FAA's CARI-7A that agrees with it to within about 1% on reference flights. It shows a yearly total against the 1 mSv public limit and can export a dose report for your records. For US pilots, whose employers are not required to track dose, that is a simple way to know your own number.

Sources

1. Cancer-Related Mortality Among US Pilots and Flight Attendants (Patel, Liu and Jena; doi:10.1001/jamainternmed.2026.3635), JAMA Internal Medicine, August 17, 2026
2. Pilots, Flight Attendants Have Greater Risk of Radiation-Related Cancer Death Than Other Professions, Harvard Medical School, August 17, 2026
3. AC 120-61B, In-Flight Radiation Exposure, FAA, November 21, 2014
4. Council Directive 2013/59/Euratom, EUR-Lex, December 5, 2013
5. Dose limits (ICRP Publication 103), ICRPaedia, ICRP
6. Evaluation of occupational exposure to ionizing radiation (UNSCEAR 2020/2021 Report, Annex D), UNSCEAR, 2022
7. Aircrew radiation exposure, CDC/NIOSH
8. Cancer in aircrew, CDC/NIOSH, reviewed June 3, 2024
9. Ionizing Radiation and Radiation Safety in Aerospace Environments, DOT/FAA/AM-21/8, FAA, March 2021
10. NASA, NOAA: Sun Reaches Maximum Phase in 11-Year Solar Cycle, NASA, October 15, 2024
11. Pilots and Flight Attendants Show Elevated Radiation-Related Cancer Mortality, The ASCO Post, August 2026
12. Pilots and Flight Attendants Face a Real Cancer Risk. Frequent Flyers Shouldn't Panic (Patel, Worsham and Jena), TIME, August 17, 2026