Standard Article

Electromagnetic Fields

Environmental Health

  1. David A. Savitz

Published Online: 15 SEP 2006

DOI: 10.1002/9780470057339.vae018

Encyclopedia of Environmetrics

Encyclopedia of Environmetrics

How to Cite

Savitz, D. A. 2006. Electromagnetic Fields. Encyclopedia of Environmetrics. 2.

Author Information

  1. University of North Carolina at Chapel Hill, NC, USA

Publication History

  1. Published Online: 15 SEP 2006

Abstract

Concerns with possible health effects of low-level exposure to extremely low-frequency electric and magnetic fields have been prominent for over 20 years. The hypothesis of particular interest is that prolonged exposure, over periods of months to years, to relatively elevated power-frequency (50 or 60 Hz) magnetic fields in the environment increases the risk of leukemia and brain cancer. Beyond the substantial public interest and policy relevance to the question of whether this ubiquitous exposure poses a threat to human health, the topic serves as a prototype of a low-level, ambiguous environmental health concern. Like the concerns with persistent organochlorines such as dichlorodiphenyltrichloroethane (DDT), polychlorinated biphenyls (PCBs), or dioxin, or with drinking water disinfection by-products, human health effects are at the margins of being detectable if present at all. Furthermore, as with many ubiquitous exposures such as indoor Radon, it is difficult to clearly identify exposure gradients: a few individuals may be highly exposed but most of the population is in a range in which discrimination of exposure levels is a challenge. Finally, because the exposure is so widespread, even a modest influence on disease risk is of public health interest. That is, public health concerns push back the boundaries of what is scientifically feasible, creating tension between the knowledge that is desired for setting policy on regulatory standards, and the often rather blunt instruments of environmental epidemiology. In the face of this pressure, statistical methods for maximizing the sensitivity and precision of risk estimates are paramount.