The only source of VLF magnetic fields which most people encounter in their daily lives is the horizontal deflection circuitry of a video display terminal or television set. Kavet and Tell (1991) give a clear and concise description of the operation of a video display terminal (VDT). Briefly, the image on a television set or VDT is generated by an electron beam, which scans the screen in a series of horizontal lines from top to bottom. The horizontal position of the electron beam is controlled by a VLF magnetic field, with a fundamental frequency of approximately 15-31 kHz. The vertical position of the beam is controlled by an ELF magnetic field, at approximately 30-80 Hz.
Haes and Fitzgerald (1995) have measured the strength of the VLF magnetic field at a distance of 50cm in front of approximately 140 VDTs by a wide variety of manufacturers. The measured fields had flux densities (at 50cm from the screen) between approximately 0.002uT and 0.1 uT. As discussed above, this is the strongest VLF field to which most people are routinely exposed; for comparison, the ELF fields produced by standard household appliances are typically in the range of 0.1uT and 10T at a distance of 50cm (Gauger, 1985).
Kavet and Tell (1991) describe fourteen epidemiological studies seeking to establish a link between VDT use and either spontaneous abortion or birth defects. Of these fourteen studies, one found a significant link between VDT use and spontaneous abortion, and one found a link between VDT use and first trimester spontaneous abortion; the first of these two was later faulted for interviewer bias in a study summarized by Haes and Fitzgerald (1995).
Kavet and Tell (1991) also describe studies at four laboratories seeking a link between exposure to VLF magnetic fields and reproductive outcome in rodents. The waveforms used in these studies were all sawtooth waveforms similar to those generated by a VDT, at frequencies of 18-20kHz. After scaling to a human equivalent exposure based on ``induced current considerations,'' the maximum field strengths in these studies were 1.25, 1.25, 5.51, and 16.7 uT respectively; in the two studies which tested multiple field strengths, no effect of field strength was reported. Kavet and Tell report that one of the laboratories found a significant link between VLF exposure (at 1.25 uT equivalent) and spontaneous abortion when measured per fetus, but they question the statistical validity of the result, because analysis results per litter were not reported.
A recent epidemiological study by Lindbohm et al. (1992) suggests that chronic exposure to the ELF fields of a VDT, rather than the VLF fields, may increase the risk of spontaneous abortion. For each of their subjects and controls who reported using a VDT during her pregnancy, Lindbohm et al. examined company records to determine the make and model of the VDT, and then measured the ELF and VLF emissions of a similar monitor. They found that VDT use did not increase the risk of spontaneous abortion, but that using a monitor with strong magnetic fields did. Using a monitor with ``strong'' ELF emissions (greater than 0.9uT) significantly increased the risk of spontaneous abortion, and the risk increased when field strength was weighted by the amount of time the subject used a monitor. Use of a monitor with strong VLF emissions (rate of change greater than 30 mT/s, corresponding to a field of 0.24uT at 20kHz) caused a non-significant increase in risk, which became significant when weighted by usage time. VLF and ELF emissions of all monitors were highly correlated (r=0.76); when VLF and ELF emissions were considered together, strong ELF emissions were found to increase risk significantly, while strong VLF emissions actually decreased risk by a non-significant amount.