Sun exposure measurements in populations.

Measuring ultraviolet (UV) exposure in human populations is a complex task. The dominant source of ultraviolet radiation for humans is sunlight, which is always a mix of both ultraviolet A (UVA, usually 320– 400 nm) and ultraviolet B (UVB, 320–280 nm). Ultraviolet C ( 280 nm) is filtered by the atmosphere and does not reach the earth’s surface. Researchers cannot separate the individual effects of the different wavelengths in humans as can be done in experimental animals. For some subpopulations, such as welders and those who use artificial UV sources for tanning, exposures other than sunlight may contribute to the total individual dose. In epidemiologic studies in which sun exposure is estimated, the usual end point is skin dose. The major acute toxicities of sun exposure are photosensitivity reactions and sunburn; chronic toxicities include melanoma, non-melanoma skin cancers, cataracts, and photo-aging of the skin. Two major methodologies for estimating sunlight exposure have emerged: “geographic” estimates and personal, individual estimates. Geographic estimates of ground-level exposure must take into account latitude, altitude, day of year, time of day, cloud cover, ozone, particulates, smog, and reflectance from water, rock, or sand. All of these cofactors have important influences on the actual ground-level measurements. Several US agencies have had on-ground measurement devices in place for varying times. The National Cancer Institute (NCI) and the National Oceanic and Atmospheric Administration (NOAA) collaborated on placing Robertson-Berger meters, spectrometers measuring on-ground UVB flux, in areas with cancer registries. These meters were in place for extended periods of time. The limitations include only UVB measurements and yearly calibration of the instruments. These data have been used in analyses of risk of non-melanoma skin cancer and melanoma. The Environmental Protection Agency (EPA) had 21 multispectral (286.5–363 nm in 0.5 nm steps) Brewer meters spread over the United States in place for several years. The system is not functional now, but data are available for use. The US Department of Agriculture (USDA) has 37 monitoring stations in 29 states, Canada, and New Zealand. These, too, are multispectral, but are focused on UVB. Most are placed in rural agricultural areas, since the major end point for these measurements is the effect of UVB on crops. This system was initiated in the mid-1990s. NOAA has six stations in the western, central, and eastern United States with Brewer meters in place since late 2006. Data are also available from these stations. In addition to these on-ground measurement systems, NASA provides Total Ozone Mapping Spectrometer (TOMS) satellite data with on-ground estimates. The advantage of the TOMS data is that there is superior coverage from 180 degrees E to 180 degrees W and 65 degrees N to 65 degrees S from 1978 –1993. The disadvantage is that the estimates cannot take into account cofactors that alter the ground measurements. One study compared TOMS estimates with on-ground measurements and found agreement within 12% overall, which improved with clear skies (within 8%). Based on measurements in two not very urban areas (Billings, OK and Las Cruces, NM), the effect of aerosols was less than 5%, but would likely be higher in more congested areas. In addition to the geographic measures, personal dosimeters have also been developed. Polysulfone badges have been used for over a decade. They are calibrated to erythemal-weighted UVB and measure total dose over a specified time period. Placement of the badges is crucial because the topography and movement of a human being determines the total exposure. A major advantage of these badges is that they are easy to use and relatively inexpensive, making them feasible for moderate-sized studies. Multispectral dosimeDr. Tucker is with the Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, Maryland, USA. Please address all correspondence to: Dr. Margaret Tucker, Director, Human Genetics Program, Chief, Genetic Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, 6120 Executive Blvd., Room 7122, Bethesda, MD 20892-7236, USA; Phone: 301-496-4375; Fax: 301402-4489; E-mail: tuckerp@mail.nih.gov. doi: 10.1301/nr.2007.aug.S84–S86