UV radiation and melanoma: WHO report and IARC data

A historical document, with its date and context

Solar Ultraviolet Radiation: Global burden of disease from solar ultraviolet radiation, published by WHO in 2006, assessed the burden of disease associated with solar radiation. For melanoma, its sources included incidence and mortality estimates for 2000. The PDF already held in the site’s archive remains available in full, but should not be read as a picture of the current situation.

This page therefore distinguishes three levels: the report’s historical methods and findings, subsequent updates and practical implications for prevention. A study’s publication date and the year represented by its data are not necessarily the same. An editorial update to this page does not change the date of the original data.

Incidence means new cases diagnosed within a period; mortality means deaths. Disease burden may also include years of life lost and years lived with disability, combined as DALYs. These indicators are related but different: a percentage of cases cannot automatically describe deaths, severity or an individual’s probability of recovery.

Read the original 2006 WHO report (PDF, English)

What population attributable fraction means

Population attributable fraction, or PAF, estimates the share of cases that would not have occurred in a reference scenario with lower exposure. It requires assumptions about the causal relationship between exposure and disease and the population used for comparison. It is a public-health measure, not a test to apply to one person’s tumour.

Saying that a proportion of melanomas is attributable to UV does not mean each such case can be identified with certainty, or that one sunscreen can prevent the same proportion of cases. The model considers an exposure history and a theoretical comparison; the effectiveness of a practical intervention requires its own evidence.

To interpret a PAF, ask which cancer is included, in which population, in which year and relative to which scenario. Cutaneous melanoma estimates should not automatically be extended to melanomas of the eye or mucosal surfaces. Even within the skin, subtypes differ in their relationship with sun exposure.

Historical estimates and why they differed

Historical material reported very high estimates from ecological comparisons: 96% in men and 92% in women in the United States, and 89% and 79% when comparing light-skinned populations in New South Wales with England and Wales. These figures belong to specific comparisons and assumptions of their period; they are not universal percentages to assign to every country today.

Case-control studies discussed in the report produced estimates around 0.2, whereas lower and upper scenarios of 0.5 and 0.9 were used to calculate disease burden. The upper value came from ecological data and the lower from expert consensus. These were not simply three interchangeable measurements of the same phenomenon.

Discussed difficulties included reconstructing lifetime exposure, recall errors and the absence of a genuinely UV-unexposed human group. Measurement errors can weaken an observed association. Conversely, comparisons between whole populations may incorporate differences in pigmentation, behaviour and access to diagnosis that are not solely related to sunlight.

The methodological implication matters: do not choose the highest number because it sounds persuasive, or the lowest because it is reassuring. Examine how it was obtained, its limitations and the question it answers. The PDF retains the details and references behind the original reasoning.

Latitude, pigmentation and geographical comparisons

The report discussed increasing incidence at lower latitudes within certain populations, but also apparent reversals when different countries were compared. Examples included differences between northern and Mediterranean Europe and an Italian pattern unlike that observed in other settings. Location on a map is therefore not a sufficient measure of risk.

In historical models discussed by Langford, the United Kingdom, Ireland, Belgium and the Netherlands showed positive relationships between mortality and ambient UVB exposure; France, Italy, Germany and Denmark showed different patterns. These observations do not demonstrate that sunlight is protective in certain countries. Incidence and mortality are affected by population composition, intermittent travel exposure, diagnosis and other factors.

The document also noted limited studies in darker-skinned populations and a greater proportion of palmoplantar melanomas among observed cases. Acral forms do not have the same relationship with UV as typically sun-associated forms. Using a group as a statistical reference does not mean its members have no risk or no need to recognise suspicious lesions.

IARC update: 2022 data published in 2025

An IARC study announced in May 2025 estimated that 83% of new cutaneous melanomas worldwide in 2022 were attributable to UV exposure. The estimated proportion was 86% in men and 79% in women. The greatest burden concerned older age groups and regions with predominantly light-skinned populations.

These are results of an epidemiological model using population data, not a test performed on every tumour. The Global Cancer Observatory method uses a theoretical minimally exposed reference; the portal explains its assumptions and geographical differences. The 83% figure cannot be converted into a personal probability or a prevention guarantee.

The new figure should not be mechanically compared with percentages from the 2006 report to claim that risk has risen or fallen by a particular amount: the year, information base and method differ. For case trends and Italian data, also see melanoma epidemiology, keeping disease frequency distinct from the proportion attributable to an exposure.

What this means for prevention

The practical message remains to reduce avoidable exposures and sunburn, without treating tanning as a health objective. The UV index helps planning; clothing, a hat, sunglasses, shade and sunscreen on uncovered skin work together. Tanning-bed exposure is not a safe alternative.

Vitamin D, also discussed in the historical document, does not justify sunburn or tanning programmes. Nutritional needs, possible deficiency and supplementation should be assessed individually with a clinician. Disease-burden estimates do not provide a personal dose of sunlight.

Protection does not remove individual risk factors or replace attention to new or changing lesions. People with previous melanoma, numerous moles or family history should follow the pathway agreed with their specialist. Reading the data is useful when it informs these choices without creating false precision about an individual’s future.