A PERSPECTIVE ON ELECTROMAGNETIC-FIELD BIOEFFECTS AND RISK ASSESSMENT

There is evidence to suspect that a human health risk could exist from exposure to power frequency electromagnetic fields (EMFs). Summary an alyses of epidemiological evidence suggest that, if the effects seen a re real, they are small, with relative risks of the order 1.5-2.5. The past several years of epidemiological results are consistent with tho se of previous years, but do not add markedly to the understanding of potential human health risk. A similar situation has existed with benz ene. Health concerns for benzene originated in the workplace with case reports. For many years, epidemiology studies alone provided the info rmation on risks from benzene; until recently, animal models did not y ield any tumors. Finally after 50 years of laboratory studies, an anim al model provided tumors, but not the primary one associated with benz ene exposure to humans. A brief history of the benzene story is presen ted to provide a perspective on the progress in understanding the pote ntial risk of EMFs. Radon and toxic chemicals, which are present in in door air, have been examined qualitatively only in a few EMF epidemiol ogical studies. Because these materials have the potential to overshad ow any signal coming from EMF exposures, they must be treated quantita tively. No classical cancer studies have been completed as yet for EMF s, but a variety of promotion or copromotion animal studies have been reported. Results of these and other studies lend support for consider ation of EMF not being a complete carcinogen, but possibly a promoter. Qualitatively, whole animal studies, together with results of epidemi ological works, suggest that greater intensities (magnetic flux densit ies) may be associated with greater responses (or lower p values). Oth er laboratory studies have been designed to examine more basic mechani sms, but the use of these types of studies in risk assessment is not s traightforward. A methodology, based on relative potency concepts, is introduced; it can be used to assist in the assessment of the degree o f hazard which EMF poses. EMF researchers can contribute to the use of this type of analysis by including positive controls in their experim ental designs.