Epidemiological study (observational study)

Residential magnetic fields predicted from wiring configurations: II. Relationships To childhood leukemia epidem.

By: Thomas DC, Bowman JD, Jiang L, Jiang F, Peters JM

Published in: Bioelectromagnetics 1999; 20 (7): 414-422

Aim of study (acc. to author)

The data of a case-control study on childhood leukemia in Los Angeles County, USA, were reanalyzed with residential magnetic fields predicted from the wire configurations of nearby power transmission lines.

Further details

The principal residence of a child was defined as the house within southern California where the child lived in for the longest portion of etiologic period, i.e., the time period from conception to 0-2 years before diagnosis of the case. If no single residence spanned at least 40 % of the etiologic period, but two homes each spanned at least 25 %, both were designated as principal residences. Measurements were carried out for 315 principal houses of 164 cases and 144 controls. Maps with the complete information about nearby transmission and distribution wiring were obtained for 709 residences. The measured and predicted fields were categorized by their 50th, 75th, and 90th percentiles.

Endpoint/type of risk estimation

childhood leukemia

Type of risk estimation:

incidence

(odds ratio (OR))

Exposure

Assessment

interview: residential history, use of electric appliances
wire code: Wertheimer-Leeper scheme
measurement: electric and magnetic fields measurement under child's bed for 24 hours
calculation: 24-h geometric mean of the magnetic field magnitude in the child's bedroom

Exposure groups

Group	Description
Group 1	24-h geometric mean measurements, measured principal: percentiles 50-74 %
Group 2	24-h geometric mean measurements, measured principal: percentiles 75-89 %
Group 3	24-h geometric mean measurements, measured principal: percentiles 90-100 %
Group 4	24-h geometric mean measurements, corrected for measurement error, measured principal: percentiles 50-74 %
Group 5	24-h geometric mean measurements, corrected for measurement error, measured principal: percentiles 75-89 %
Group 6	24-h geometric mean measurements, corrected for measurement error, measured principal: percentiles 90-100 %
Group 7	Wertheimer-Leeper code, measured principal: ordinary low (OLCC)
Group 8	Wertheimer-Leeper code, measured principal: ordinary high (OHCC)
Group 9	Wertheimer-Leeper code, measured principal: very high (VHCC)
Group 10	Wertheimer-Leeper code, all principal: ordinary low (OLCC)
Group 11	Wertheimer-Leeper code, all principal: ordinary high (OHCC)
Group 12	Wertheimer-Leeper code, all principal: very high (VHCC)
Group 13	predicted fields, measured principals: percentiles 50-74 %
Group 14	predicted fields, measured principals: percentiles 75-89 %
Group 15	predicted fields, measured principals: percentiles 90-100 %
Group 16	predicted fields, all principals: percentiles 50-74 %
Group 17	predicted fields, all principals: percentiles 75-89 %
Group 18	predicted fields, all principals: percentiles 90-100 %
Group 19	predicted fields, all mapped: percentiles 50-74 %
Group 20	predicted fields, all mapped: percentiles 75-89 %
Group 21	predicted fields, all mapped: percentiles 90-100 %
Group 22	predicted fields, all residences: percentiles 50-74 %
Group 23	predicted fields, all residences: percentiles 75-89 %
Group 24	predicted fields, all residences: percentiles 90-100 %
Group 25	measurements and predictions combined: measured principal: percentiles 50-74 %
Group 26	measurements and predictions combined: measured principal: percentiles 75-89 %
Group 27	measurements and predictions combined: measured principal: percentiles 90-100 %
Group 28	measurements and predictions combined: all principal: percentiles 50-74 %
Group 29	measurements and predictions combined: all principal: percentiles 75-89 %
Group 30	measurements and predictions combined: all principal: percentiles 90-100 %
Group 31	measurements and predictions combined: all mapped: percentiles 50-74 %
Group 32	measurements and predictions combined: all mapped: percentiles 75-89 %
Group 33	measurements and predictions combined: all mapped: percentiles 90-100 %
Group 34	measurements and predictions combined: all residences: percentiles 50-74 %
Group 35	measurements and predictions combined: all residences: percentiles 75-89 %
Group 36	measurements and predictions combined: all residences: percentiles 90-100 %

Population

Group:
- children
Age: 0–10 years
Observation period: 1980 - 1987
Study location: USA (Los Angeles County, California)

Case group

Characteristics: leukemia
Data source: Los Angeles County Cancer Surveillance Program

Control group

Selection:
- friends of cases or randomly digit dialing
Matching:
- sex
- age
- ethnicity

Study size

	Cases	Controls
Eligible	331	257
Participants	232	232

Statistical analysis method:

logistic regression
trend test

Results (acc. to author)

No association between measured magnetic fields and childhood leukemia was observed, but the risks were significant for predicted magnetic fields above 0.125 µT (group 24: OR 2.0, CI 1.03-3.89, number of cases and controls not specified), and a significant dose response was seen (P for trend = .02). When exposure assessment was done by a combination of predictions and measurements with corrections for errors, the odds ratio (group 36: OR 2.19, CI 1.12-4.31) and the trend (p = .007) showed somewhat greater significance.
These findings support the hypothesis that magnetic fields from electrical lines are causally related to childhood leukemia but that this association has been inconsistent among epidemiologic studies due to different types of errors in exposure assessment.

Study funded by

Los Angeles Department of Water and Power (LADWP), USA
Southern California Edison (SCE), USA

Comments on this article

Thomas DC (2018): Re: Kheifets et al. (2017): Residential magnetic fields exposure and childhood leukemia: a population-based case-control study in California

Seomun G et al. (2021): Exposure to extremely low-frequency magnetic fields and childhood cancer: A systematic review and meta-analysis
Auger N et al. (2019): Residential exposure to electromagnetic fields during pregnancy and risk of child cancer: A longitudinal cohort study
Crespi CM et al. (2019): Childhood leukemia risk in the California Power Line Study: Magnetic fields versus distance from power lines
Amoon AT et al. (2018): Proximity to overhead power lines and childhood leukaemia: an international pooled analysis
Swanson J et al. (2018): Reanalysis of risks of childhood leukaemia with distance from overhead power lines in the UK
Kheifets L et al. (2017): Residential magnetic fields exposure and childhood leukemia: a population-based case-control study in California
Crespi CM et al. (2016): Childhood leukaemia and distance from power lines in California: a population-based case-control study
Pedersen C et al. (2015): Residential exposure to extremely low-frequency magnetic fields and risk of childhood leukaemia, CNS tumour and lymphoma in Denmark
Linet MS et al. (1997): Residential exposure to magnetic fields and acute lymphoblastic leukemia in children
Feychting M et al. (1993): Magnetic fields and cancer in children residing near Swedish high-voltage power lines
Olsen J et al. (1993): Residence near high voltage facilities and risk of cancer in children
London SJ et al. (1991): Exposure to residential electric and magnetic fields and risk of childhood leukemia
Coleman MP et al. (1989): Leukaemia and residence near electricity transmission equipment: a case-control study
Savitz DA et al. (1988): Case-control study of childhood cancer and exposure to 60-Hz magnetic fields
Tomenius L (1986): 50-Hz electromagnetic environment and the incidence of childhood tumors in Stockholm County
Fulton JP et al. (1980): Electrical wiring configurations and childhood leukemia in Rhode Island
Wertheimer N et al. (1979): Electrical wiring configurations and childhood cancer