Study type: Epidemiological study (observational study)

A pooled analysis of magnetic fields and childhood leukaemia epidem.

Published in: Br J Cancer 2000; 83 (5): 692-698

Aim of study (acc. to author)

The association between exposure to 50-60 Hz magnetic fields and childhood leukaemia was investigated. For this purpose a pooled analysis was conducted based on primary data from nine studies carried out in Canada (McBride et al.), Denmark (Olsen et al.), Finland (Verkasalo et al.), Germany (Michaelis et al. ), New Zealand (Dockerty et al.), Norway (Tynes et al.), Sweden (Feychting et al.), USA (Linet et al.), and UK (UK Childhood Cancer Study Investigators).
Second, it was analysed whether the adjustment for the confounders socioeconomic status, mobility, level of urbanization, detached/not-detached dwelling, and level of traffic exhaust change the results.
Third, the so-called wire-code paradox (a stronger association between proxy measures of magnetic field and cancer than between direct measurements and cancer) was investigated on the combined data.

Further details

All recent studies were included in which the exposure was assessed by 24 hour or 48 hour magnetic field measurements or calculated fields. For each case (correspondingly for the controls) the average exposure during the last year prior diagnosis was calculated based on the primary data to achieve data being as consistent as possible across the studies.

Endpoint/type of risk estimation

Type of risk estimation: (relative risk (RR))

Exposure

Assessment

Exposure groups

Group Description
Group 1 geometric mean of the measured magnetic fields: < 0.1 µT
Group 2 geometric mean of the measured magnetic fields: 0.1 - 0.2 µT
Group 3 geometric mean of the measured magnetic fields: 0.2 - 0.4 µT
Group 4 geometric mean of the measured magnetic fields: ≥ 0.4 µT
Group 5 calculated magnetic fields: < 0.1 µT
Group 6 calculated magnetic fields: 0.1 - 0.2 µT
Group 7 calculated magnetic fields: 0.2 - 0.4 µT
Group 8 calculated magnetic fields: ≥ 0.4 µT
Group 9 measured and calculated magnetic fields: < 0.1 µT
Group 10 measured and calculated magnetic fields: 0.1 - 0.2 µT
Group 11 measured and calculated magnetic fields: 0.2 - 0.4 µT
Group 12 measured and calculated magnetic fields: ≥ 0.4 µT
Reference group 13 Wertheimer-Leeper configuration: underground + very low
Group 14 Wertheimer-Leeper configuration: ordinary low
Group 15 Wertheimer-Leeper configuration: ordinary high
Group 16 Wertheimer-Leeper configuration: very high

Population

Study size

Type Value
Total 3,247
Other:

2704 (83 %) cases of acute lymphatic leukaemia

Statistical analysis method: (adjustment: )

Results (acc. to author)

Only 0.8 % of the children were exposed to estimated magnetic fields of 0.4 µT and above. No evidence was found for an increased risk of childhood leukemia at residential magnetic field levels below 0.4 µT. A statistically significant relative risk estimate of 2 for childhood leukemia was observed for children with residential magnetic fields of 0.4 µT and more. The results did not change following adjustment for the potential confounders. The authors found no evidence for the existence of the so-called wire-code paradox.

Study funded by

Comments on this article

Related articles