Purpose: A meta-analysis was conducted to obtain a 'quantitative' estimate of the extent of genetic damage in mammalian somatic cells exposed to non-ionizing radiation emitted from extremely low frequency electro-magnetic fields (ELF-EMF) and to compare with that in unexposed control cells.
Methods: The methods used for the meta-analysis were recommended in several standard text books. Three specific variables related to ELF-EMF exposure characteristics were examined in the meta-analysis: (i) frequency (Hz), (ii) flux density (mT), and (iii) in occupationally exposed individuals.
Result and conclusions: (1) The difference between ELF-EMF-exposed and control cells as well as the 'effect size' due to ELF-EMF exposure were biologically small (although statistically significant) with very few exceptions. (2) At certain ELF-EMF exposure conditions there was a statistically significant increase in genetic damage assessed from some end-points. (3) The mean indices for chromosomal aberrations and micronuclei end-points in ELF-EMF-exposed and control cells were within the spontaneous levels reported in historical database. (4) Considerable evidence for publication bias was found in the meta-analysis.