Do studies of wire code and childhood leukemia point towards or away from magnetic fields as the causal agent?

A long-standing point of controversy in the epidemiologic literature concerns the meaning of a wire code-childhood leukemia association for assessing the role of magnetic field exposure. Six studies of wire codes and childhood leukemia in North America were examined, three of which reported positive associations and all of which found some relation between wire codes and measured magnetic fields. Supporting magnetic fields as the basis for the wire code associations are the correspondence between those wire code levels which predict distinct magnetic fields and those which predict leukemia risk in the positive studies. Geographic locations and methods that refine wire codes as magnetic fields predictors also tend to strengthen the association with leukemia. Opposing arguments are based on the failure of the wire code-magnetic field association to predict the strength of association across studies, including the unexplained lack of association between wire codes and leukemia in the Midwest and in Canada. Alternatives to magnetic fields are less supported; residential mobility, social class, and neighborhood characteristics are unlikely to explain a wire code effect. Ambiguity persists because of the modest strength of the wire code-leukemia association, the complexity of the relation between wire codes and magnetic fields, lack of knowledge of risk factors for childhood leukemia, and the limited evaluation of wire code correlates other than magnetic fields.