Exposure Science Issues Concerning 60 Hz Magnetic Fields.

Abstract

Several recent epidemiology studies suggest that exposure to magnetic fields may be one of the etiologic factors involved in adverse reproductive health outcomes, but these studies potentially had several important design limitations that undermine the validity of their findings and subsequent conclusions. This research examined these limitations in detail using hypothesis-driven data collection and statistical analyses with the underlying goal of informing the design of future epidemiology studies concerning exposure to magnetic fields and adverse reproductive health. The study design and other related exposure science issues examined by this research included: 1) the adequacy of using a single day’s worth of personal magnetic field exposure data to characterize longer periods of exposure; 2) the potential influence of physical activity on personal magnetic field exposure; and 3) the comparison of personal magnetic field exposures between women and men and within female-male couples. These issues were assessed with data from two longitudinal cohorts of men and/or women recruited from prenatal care clinics in North Carolina and an infertility center in Massachusetts. We observed that measures of central tendency associated with daily personal magnetic field exposures were more stable over time compared with measures of peak, and the stability of these metrics was greater over short- relative to long-term durations. The findings suggest that if there is interest in peak exposure metrics, more than one day of measurement is needed over the window of disease susceptibility to reduce measurement error. We also observed a positive relationship between physical activity and peak magnetic field exposure metrics, suggesting physical activity could be an important confounder in the relationship with any outcome independently associated with activity, such as miscarriage, and, as a result, should be adjusted for in statistical models to reduce bias. In addition, we demonstrated that distributions of personal exposures among women and men are similar, and that there is promise that one partner’s exposure data could be used as a surrogate for the other’s in the absence of such data. Future reproductive health epidemiology studies that concern exposure to magnetic fields should consider this research in the design and interpretation of their findings.