Flight attendants and cancer risk: what the studies actually show

The question "is being a flight attendant bad for you" is a real epidemiological question with a real literature. The picture that emerges from the better-designed studies is that cabin crew have elevated incidence of certain cancers relative to the general population, that the magnitudes are modest, and that cosmic radiation is one of several plausible contributing exposures along with circadian disruption from frequent time-zone changes and (historically) secondhand cabin smoke. This guide summarises what the literature actually supports, what it does not, and how to read findings about cabin-crew health.

The headline US study: NIOSH Flight Attendant Health Study

The most-cited US data come from the National Institute for Occupational Safety and Health Flight Attendant Health Study, a cohort study of more than 5,000 US cabin crew followed for cancer incidence and other health outcomes. The principal published finding is that flight attendants had elevated standardised incidence ratios (SIR — observed cancers ÷ expected cancers based on general-population rates) for several cancer types ^[1]:

Cancer type	Approx SIR (FAs vs general US population)	Plausible contributing exposures
Breast cancer (female FAs)	~ 1.5	Cosmic radiation, circadian disruption, reproductive-history factors
Melanoma	~ 2.0	UV exposure (FAs travel to high-UV locations off-duty), possibly cosmic UV at altitude
Non-melanoma skin cancer	~ 4.0	UV exposure (same factors)
Thyroid cancer	~ 1.2	Cosmic radiation; ascertainment bias (more medical contact)
Gastrointestinal cancers (combined)	~ 1.2	Multiple — none specific

For most cancer types not listed, FAs showed incidence comparable to or below the general US population. Notably, lung cancer incidence in flight attendants is lower than the general population, reflecting historically lower smoking rates in the FA cohort (notwithstanding past secondhand smoke exposure during the indoor-smoking era).

Earlier European cohorts

European cohort studies of cabin crew (Nordic countries, Germany, Italy) have produced broadly consistent findings — elevated breast cancer, elevated melanoma — with effect sizes in similar ranges ^[2]. The European data also suggest a dose-response relationship between cumulative flight hours and breast cancer risk, though confounding by reproductive-history variables (parity, age at first birth) is difficult to fully untangle.

What does "elevated by 50%" actually mean for an individual

A standardised incidence ratio of 1.5 for breast cancer in female FAs means that, statistically, the FA cohort experienced 50% more breast cancer cases than would be expected for a general-population cohort matched on age and (where adjusted) other factors. In absolute terms, the US female lifetime risk of breast cancer is approximately 13%. A 50% relative-risk elevation pushes that to approximately 19% — meaningful, but not transformative.

For melanoma the absolute lifetime risk in the US population is roughly 2.5%; a 2× elevation pushes it to 5%. For non-melanoma skin cancer the absolute risk is high to begin with (most cases are basal-cell carcinomas, very treatable) and a 4× SIR represents a significant occupational excess but a manageable one.

Attribution to cosmic radiation specifically

Here the literature is honest about its limits. Cosmic radiation is a plausible contributor to:

• Breast cancer — though the dose received is low relative to known breast-cancer-causing radiation doses (the BEIR VII coefficient for breast cancer at low doses is non-trivial but small).
• Thyroid cancer — the thyroid is somewhat radiosensitive; the elevation in FAs is small.

Cosmic radiation is a less plausible contributor to:

• Melanoma and non-melanoma skin cancer — these are overwhelmingly UV-driven, and FAs have above-average UV exposure for non-occupational reasons (frequent travel to sunny destinations).
• GI cancers, lung cancers — radiation is a minor risk factor compared to dietary, smoking, and other lifestyle exposures.

Several review papers (notably the IARC monographs and ICRP-132's discussion of the epidemiological evidence) conclude that the cancer excess in cabin crew is consistent with the combined contributions of cosmic radiation, circadian disruption, and (historically) secondhand smoke, without being able to apportion the relative contributions precisely ^{[3, 4]}.

Circadian disruption is a real second exposure

IARC classified shift work involving circadian disruption as Group 2A — "probably carcinogenic to humans" — in 2007, based on multiple lines of evidence including animal studies and epidemiology of female night-shift workers ^[3]. Long-haul cabin crew have one of the most extreme circadian-disruption work patterns of any occupation. Disentangling the cancer-incidence contribution of circadian disruption from cosmic radiation is genuinely difficult, and an honest reading of the literature is that both are likely contributing in some proportion.

What the studies do not show

• They do not show that flight attendants have shorter lifespans than the general population. Several cohorts show longer all-cause life expectancy, consistent with a healthy-worker effect.
• They do not show consistent excess in leukaemia or in cancers strongly associated with high radiation doses in atomic-bomb-survivor cohorts. This is important: it suggests cosmic-radiation effect, if present, is modest.
• They do not establish a quantitative cumulative-dose-response relationship sharp enough to derive individual risk coefficients beyond what BEIR VII and ICRP already provide.

What this means for an individual

If you are aircrew: the occupational excess of certain cancers is real but the absolute increase is modest, and the dose-mitigation pathways (lower-latitude assignments, schedule modification, pregnancy accommodation) are exactly what the regulatory framework (EURATOM in the EU, FAA AC 120-61B in the US) is designed to support. Standard cancer screening recommendations for your age and sex apply; some commentators argue for slightly earlier or more frequent breast screening for long-haul female aircrew, though this is a clinical judgement not yet incorporated into general guidelines.

If you are a frequent flier passenger: your occupational profile differs substantially from cabin crew. You typically have lower cumulative flight hours, much less circadian disruption from on-aircraft work patterns, and no past occupational exposure to secondhand smoke. The cabin-crew studies are not directly applicable to you. The relevant numbers for you are the dose figures in our frequent-flyer math guide, contextualised by the BEIR VII risk coefficients applied to your cumulative lifetime dose.

How to read future studies

When you see a headline about a new cabin-crew cancer study, the questions to ask:

1. What is the cohort? Cabin crew at one airline, in one country, over what years?
2. Were results adjusted for reproductive history, smoking, and BMI?
3. Was cumulative flight-hour data available, or only employment duration?
4. Are the absolute risk differences clinically meaningful, or only the relative risks?
5. Is the cosmic-radiation contribution separated from circadian-disruption contribution, or treated as a single occupational-exposure bucket?

Well-designed studies report the answers to these questions in the methods section. Less-careful press coverage tends to lose them in translation.

The healthy-worker effect

Almost every occupational cohort study runs into the healthy-worker effect — the population of people fit and motivated enough to be hired and retained in a demanding job is, on average, healthier than the general population at baseline. The healthy-worker effect tends to suppress observed mortality and disease rates in occupational cohorts and is one reason all-cause mortality figures for cabin crew tend to look favourable. Cancer-incidence figures are less affected because cancer is generally weakly correlated with the variables (fitness, smoking) that drive the healthy-worker effect in cardiovascular and respiratory disease.

For a careful reading of the NIOSH and European cabin-crew studies, the standardised incidence ratios for cancers like breast cancer and melanoma should be interpreted as conservative — that is, they may understate the true occupational excess because the comparison group includes people who would not pass the medical and occupational requirements to be cabin crew in the first place.

What the literature does and does not separate

Most of the cohort studies treat "flight attendant" as a single occupational exposure category. They typically have data on years employed but limited data on cumulative flight hours, route mix (short-haul vs long-haul, polar vs equatorial), or specific time-zone-crossing patterns. As a result the studies can identify that there is excess cancer in the cohort but cannot cleanly identify which aspect of the work is responsible.

The best-designed studies attempt some stratification — for example, splitting the cohort into long-haul vs short-haul flying or into pre-1990 vs post-1990 employment to control for the secondhand-smoke transition. These finer-grained analyses generally find that long-haul flying carries a larger excess, consistent with both higher cosmic-radiation exposure and more circadian disruption. Disentangling those two contributions individually is, with current data, beyond the resolution of the studies.

What is changing in the modern era

The flight-attendant cohorts most cited in the literature span careers that began in the 1960s through 1990s. Several things have changed for current and future cohorts that may make the historical findings imperfect predictors of current risk:

• Indoor smoking ban. Cabin smoking was eliminated on US domestic flights in 1990, on international flights through the mid-2000s. Current cabin crew have essentially no occupational secondhand-smoke exposure, removing a significant historical contributor to lung-cancer risk.
• Aircraft cabin altitude. The 787 and A350 maintain lower cabin altitudes than the older 747 and 767. This reduces some non-radiation occupational stressors but does not affect cosmic-radiation dose.
• Flight-pattern intensity. Long-haul polar flying has expanded substantially since the late 1990s, particularly for crew on cross-polar Asia-North America routes. Current long-haul cohorts may receive somewhat higher per-year doses than the cohorts studied in the historical literature.
• Better dosimetry tracking. EU EURATOM and FAA AC 120-61B frameworks have improved data quality on per-crew cumulative dose, making future cohort studies more powerful than the historical ones.

The expectation is that future cohort studies will report somewhat different findings — likely with reduced lung-cancer excess (no more secondhand smoke), possibly slightly increased breast-cancer or thyroid excess if long-haul polar exposure continues to grow. The honest answer is that current data are based on yesterday's flying patterns, not today's.

Last reviewed 30 June 2026 · See our methodology and sources.