Wire codes, magnetic fields, and childhood cancer.

Childhood cancer has been modestly associated with wire codes, an exposure surrogate for power frequency magnetic fields, but less consistently with measured fields. We analyzed data on the population distribution of wire codes and their relationship with several measured magnetic field metrics. In a given geographic area, there is a marked trend for decreased prevalence from low to high wire code categories, but there are differences between areas. For average measured fields, there is a positive relationship between the mean of the distributions and wire codes but a large overlap among the categories. Better discrimination is obtained for the extremes of the measurement values when comparing the highest and the lowest wire code categories. Instability of measurements, intermittent fields, or other exposure conditions do not appear to provide a viable explanation for the difference between wire codes and magnetic with respect to the strength and consistency of their respective association with childhood cancer.