CIESIN Thematic Guides

Risk Factors that Influence the Occurrence of Skin Cancer within Human Populations

Researchers are examining the relationship of the growing risk of skin cancer to increases in ground-level UV-B radiation due to ozone depletion. Strong evidence exists of a dose-response relationship between nonmelanoma skin cancer and cumulative exposure to UV-B radiation. Increased risk of malignant melanoma is associated with episodes of acute exposure that result in severe sunburns, especially those that occur during childhood. In general, the incidence of nonmelanoma and malignant melanoma skin cancer has increased significantly over the past few decades, particularly in the United States, Canada, Australia, the United Kingdom, and Scandinavian countries. Behavioral changes, such as the popularity of sunbathing, over the past few decades are also considered key risk factors. In the past behavioral changes may have been the primary reason for the rising incidence of skin cancer; however, given the current trend in outdoor activities, increases in ground-level UV-B radiation may raise exposure levels even higher resulting in an even greater risk of skin cancer. Case studies have already identified high risk populations in terms of demographic, geographic, physical, and behavioral characteristics.

In addition to exposure to ultraviolet radiation, other risk factors influence the occurrence of skin cancer within human populations. Two papers presented here consider the overall epidemiology of skin cancer in terms of high risk populations. In "Epidemiology of Malignant Melanoma," Dubin, Moseson, and Pasternack (1986) report the results of a case-control study started in 1979 that identified factors related to the high incidence of malignant melanoma among Caucasians, including pigmentary traits, sun exposure, and family history. In the 1989 book Ozone Depletion: Health and Environmental Consequences, Elwood's chapter "Epidemiology of Melanoma" focuses on melanomas that arise from skin melanocytes. Elwood points out that predicting how stratospheric ozone depletion will affect the incidence of melanoma is very difficult, because the quantitative relationship between sun exposure and melanoma risk is itself complicated. Risk variables, in addition to solar dose, include the regularity and timing of exposure, the pigmentation of the individual, and past exposures.

A major publication on this subject is the 1987 U.S. Environmental Protection Agency report Ultraviolet Radiation and Melanoma (Longstreth 1987). This weighty document includes 19 chapters and two appendices along with a multitude of tables and figures covering introductory material, a review of epidemiological information, a review of experimental evidence, and dose-response relationships and conclusions. Given space constraints, only a few of the chapters from the epidemiological section are available here: Chapter 4 "Time-Related Factors in the Incidence and Mortality: Age, Period, and Birth Cohort Effects"; Chapter 6 "Geographic Distribution"; Chapter 7 "Migrant Studies"; Chapter 8 "Correlations with Indicators of Intermittent or Severe Sun Exposure"; Chapter 9 "Correlations with Measures of Cumulative Sun Exposure"; Chapter 10 "Skin Pigmentation as a Risk Factor"; Chapter 11 "Correlations with Socio-Economic Status and Occupational Factors"; and Chapter 12 "Other Factors: Steroid Hormones and Malignnt Melanoma".

Racial characteristics, principally the variation in skin pigmentation, appear to be significant in the incidence of skin cancer. The data from 59 population-based cancer registries around the world are presented in Crombie's 1979 paper "Racial Differences in Melanoma Incidence." His findings support the hypothesis that the occurrence of malignant melanoma is directly related to UV radiation exposure. Light-skinned individuals are at greatest risk, and those with darker complexions are protected to varying degrees by their melanin pigmentation. In addition, skin cancer has also been associated in some studies with hair and eye color, ancestry, and family history of the disease. In "Pigmentary Traits, Ethnic Origin, Benign Nevi, and Family History as Risk Factors for Cutaneous Malignant Melanoma," Holman and Armstrong (1984) quantify the relative and independent contributions of a wide range of demographic characteristics and environmental factors in the incidence of malignant melanoma based on a case-control study of a Caucasian population in Western Australia. Kollias and Baqer (1986) examine the role pigment plays in photoprotection in "The Role of Native Pigment in Providing Protection Against UV-B Damage in Humans." The study, conducted in Kuwait, compares the pigment levels in a sample population with the minimum dose of UV-B necessary to cause erythema.

In "Malignant Melanoma: Social Status and Outdoor Work," Lee and Strickland (1980) use a combination of data from the late 1960s on the mortality of malignant melanoma by occupation and find a gradient in the risk of skin cancer from high among upper socioeconomic, professional workers to low in low status, unskilled workers for whites in the United States, England, and Wales. Outdoor workers did not exhibit any excess risk. Gallagher et al. (1987) refer to several papers that suggest elevated risk for individuals with high socioeconomic status and indoor professional occupations. These authors use data from the Western Canada Melanoma case-control study to examine further the association of socioeconomic status and its interaction with other demographic characteristics and sunlight exposure with the risk of malignant melanoma. In their 1987 paper "Socioeconomic Status, Sunlight Exposure, and Risk of Malignant Melanoma," the authors find that controlling for other confounding factors such as skin and hair color, ethnic origin, skin reaction to sunlight, and vacation and recreational exposures to sunlight substantially reduces the socioeconomic risk gradient. Similarly, controlling for skin color and sunlight exposure significantly diminishes the inverse relationship between farming and risk of skin cancer. They conclude that the appearance of a socioeconomic gradient for melanoma may in fact be due to a combination of the subjects' skin color and exposure to sunlight. Armstrong tests the "intermittent exposure hypothesis" that the pattern of exposure to sunlight could affect the incidence of malignant melanoma in the 1986 paper "Sunlight and Malignant Melanoma in Western Australia."

The Western Canada Melanoma Study also provides the data for a broader set of findings on determinants of increased risk of skin cancer. Gallagher, Elwood, and Hill (1986) summarize the results in "Risk Factors for Cutaneous Malignant Melanoma." The authors consider the role of host pigmentation, freckling and sunburn in childhood, skin reaction to sun, the combination of sunburn, suntan, and skin reaction to sun, sunlight (ultraviolet) exposure, female oral contraceptive use, smoking, dietary and alcohol consumption, viral diseases, and other chronic diseases and surgery. In a more recent paper "The Role of Childhood Exposure to Sunlight in the Development of Solar Keratoses and Non-melanocytic Skin Cancer," Marks et al. (1990) focus specifically on the degree to which age at time of exposure influences the incidence of nonmelanoma. The authors compare the proportion of persons with solar keratoses, precursors of squamous-cell carcinomas, among native-born and British migrants in Australia. They conclude that exposure to sunlight at an age under 20 is important for the development of solar keratoses and that a certain exposure time and dose is required before the formation of lesions will occur.

The growing incidence of skin cancer is of particular concern in Australia. In "Patterns of Treated Non-melanoma Skin Cancer in Queensland," Stenbeck et al. (1990) describe the results of a 1984 survey of the regional incidence of treated non-melanoma skin cancer in four major population centers of Queensland. They note age and gender differences but emphasize the high levels throughout the population. Another study of a Queensland community focuses on a broader range of risk determinants. In "Incidence and Determinants of Skin Cancer in a High-Risk Australian Population," Green and Battistutta (1990) survey the incidence of basal-cell carcinoma and squamous-cell carcinoma and consider possible risk factors such as gender, complexion, outdoor behavior, and signs of solar skin damage for a two-year period from 1985 through 1987. Their paper also provides a good summary of related survey data from the United States and Australia.