Calculated residential exposure to power frequency magnetic fields for an epidemiological study in France and comparison to measurements

• Published in: Journal of radiological protection Vol. 43; no. 2; pp. 21507 - 21521
• Main Authors: Deschamps, François, Deambrogio, Vincent
• Format: Journal Article
• Language: English
• Published: England IOP Publishing 01.06.2023
• Subjects: Child; electrical network; Electromagnetic Fields - adverse effects; Environmental Exposure - analysis; Epidemiologic Studies; epidemiology; exposure assessment; France - epidemiology; high voltage; Humans; Leukemia - epidemiology; magnetic field; Magnetic Fields; measurements; France; electrical network; epidemiology; high voltage; magnetic field; exposure assessment; measurements
• ISSN: 0952-4746; 1361-6498; 1361-6498
• DOI: 10.1088/1361-6498/acd0b9

The GEOCAP epidemiological study of paediatric cancers in relation with various environmental factors is currently being carried out in France. Proximity to high voltage overhead lines (63–400 kV) related to exposure to extremely low frequency magnetic field (ELF-MF) is one of studied factors. The study covers the mainland French territory and includes 4174 leukaemia cases and 45 000 controls recruited from 2002 to 2010. Residential exposure was calculated for 1124 subjects whose distance to power lines was small enough for their residential exposure to possibly exceed the background reference level 0.1 μ T. Residential exposure was calculated by modelling all the influencing spans close to the subjects’ residences. At final, 371 subjects out of 1124 have a calculated exposure over 0.1 μ T, including 123 subjects over 0.4 μ T, i.e. 0.25% of the study population. These 371 calculated exposures exceeding 0.1 μ T have been matched to an existing database of ELF-MF measurements performed in inhabited areas crossed by power lines, and measurements were found for 206 subjects. Comparison was possible for 156 of them. After extrapolation of the measured values to match to calculation conditions, the difference between calculations and adjusted measurements was lower than 30% for 124 subjects (80%), with no clear tendency for over/under estimation. Differences higher than 30% have been analysed and explained.