To the Editor:
In his response to our Letter, David Savitz considers the measure of maximum magnetic fields (MMF) untrustworthy because he speculates that measuring MMF is like measuring “a random event.”1 An implicit assumption in his speculation is that MMF is highly associated with daily activities and daily activities are essentially random events. While we would agree that MMF is probably more likely than TWA (time-weighted average) to be related to daily activities, daily activities are by no means “random events.” Just think of our own daily life. Most of us conduct our daily life on a routine basis, at least during weekdays. We probably do the same things in the morning, go to work through the same routes, conduct our work in the same environment, run the same errands, go back home again through the same routes. There is a great deal of repetitiveness in our daily activities. With regard to the poor correlation between two MMF measures that were obtained some 20 weeks apart in the study by Lee et al., 2 it may have had less to do with the randomness of MMF measurements, as Savitz was implying, and more to do with the possibility that a significant portion of women were measured on non-typical days. If one of the measurements was collected on a non-typical day, a poor correlation between these two measures would be expected. Although Lee et al.2 did not have information on whether a measure was obtained on a typical or non-typical day during a participant’s pregnancy, the prospective study by Li et al.3 showed that more than one third (35%) of participants in their study were measured on a non-typical day. The latter study also showed that no association was found if the study was restricted to women who were measured on a non-typical day. This suggests that peak exposures need to occur on a daily basis (or at least for most of the days) in early pregnancy to have an effect.
Of course, the ultimate resolution to the question of reliability of MMF measurement will come from studies that measure MMF on multiple days, both typical and atypical, during early pregnancy to demonstrate the stability of MMF exposure measurements. We also think that studies are needed to identify the sources that produce these peaks and whether they are characterized by unusual field characteristics or contact currents. These studies may provide useful tools for replicating epidemiological studies.
Raymond Richard Neutra
1. Savitz DA. Magnetic fields and miscarriage. Epidemiology 2002; 13: 238.
2. Lee GM, Neutra RR, Hristova L, Yost M, Hiatt RA. A nested case-control study of residential and personal magnetic field measures and miscarriages. Epidemiology 2002; 13: 21–31.
3. Li D-K, Odouli R, Wi S, et al
. A population-based prospective cohort study of personal exposure to magnetic fields during pregnancy and the risk of miscarriage. Epidemiology 2002; 13: 9–20.