Studies linking residential exposure to magnetic fields from power lines with an increased risk of childhood cancers have produced conflicting results, and their epidemiologic methods have been questioned. This inconsistency has led to uncertainty and public concern about the relationship between residential exposure to magnetic fields and the occurrence of childhood leukemia. Linet and associates evaluated this relationship in a study of children under the age of 15 years with acute lymphoblastic leukemia who were registered with the Children's Cancer Group.

The study evaluated 638 children with acute lymphoblastic leukemia and matched them with 620 control subjects. Exposure to magnetic fields generated by nearby power lines was measured. Obtaining historic data was important in tracking exposure, so mothers of the study subjects and the control subjects were asked to provide lifetime residential histories for the children. Magnetic fields were measured in homes in which the children had lived for at least six months and usually within 24 months of the diagnosis of acute lymphoblastic leukemia in affected children. Exposure was calculated as the average of the summary level for the eligible measured home, weighted by the amount of time in the residence. Four exposure categories were identified, ranging from less than 0.065 μT to 0.200 μT or more.

The relationship between power line configuration and the strength of the magnetic field varies geographically. Therefore, residential wire-code categories were used to evaluate pairs of children with acute lymphoblastic leukemia and control subjects. Measured magnetic fields increase with increasing wire-code categories. For each child, one or two homes were measured, provided that the child had lived in one of them for at least 70 percent of the reference period. Twelve of the 428 pairs of children were excluded because technicians could not locate homes or accurately diagram nearby power lines. Homes were also evaluated if mothers had lived there during at least five months of the pregnancy that produced the child with leukemia.

The authors found there was no significant excess risk of childhood acute lymphoblastic leukemia at residential magnetic field levels of 0.200 μT or greater. No significant dose-response trends were identified. There was no association with high wire codes for either the child's main residence or the mother's residence during pregnancy.

The authors conclude that data from this study do not support an increased risk of acute lymphoblastic leukemia associated with residential magnetic field exposure up to 0.200 μT. They note that there is little support for the hypothesis that high levels of residential exposure or living in homes close to electric power lines is related to childhood acute lymphoblastic leukemia.