About Us
Contacts
Company Tour
Products
Non thermal studies
Our Research
Magnetotherapy
Electro Pollution
Endogenous Fields
Media File
The Coghill Challenge
The Natural World
Origins of Electricity
EMF Video Archives
TETRA
Photon Platinum
Review of 2003 NRPB Consultation Document
BIMT Prospectus
How to find us
|
||
| Flawed epidemiology? The case for an electric field metric in childhood cancer studies. | ||
|
Roger W. Coghill (a) and Alasdair Philips (b) We have reason to believe that the power utilities and others have deliberately steered research away from the ELF electric component of the wave, despite evidence from all main categories of science that the electric rather than the magnetic component is the active parameter causing adverse biological effects. This article sets out our support for that view. Studies of claimed associations between childhood leukaemia, mainly acute lymphoblastic leukaemia (ALL), and chronic exposure to power frequency electromagnetic fields have given rise to the so-called "wire code paradox", where surrogate measures such as distance from cabling or thickness of wires have consistently shown a much stronger association than actual measurements of magnetic field strength. The electromagnetic wave from powerlines and appliances comprises both a magnetic and an electric field component, and at user distances there is no fixed relation between them, since the exposee is in the near field of a wave some 5-6000 kilometres in length. By paying epidemiological attention to wire codes the effects of the ELF electric field are inevitably included, whereas with AC magnetic field studies it is not. This fundamentally explains why the association with childhood cancers is seen in wire code studies but not so evidently in measured magnetic field studies. Electric fields are wrongly dismissed and hence not included by authors of major EMF studies, who offer shallow and erroneous explanations for their exclusion. Moreover electric field probes are difficult to incorporate in the meters provided by EPRI for many epidemiological studies, and one has to ask if this has been engineered deliberately. Finally where electric fields are reluctantly included, the issue is avoided by using exposure periods (e.g. 3 mins spot measurements, 48 hrs bedplace measurements) which are unrepresentative of actual exposure. In their analysis of the large National Cancer Institute Childrens’ Cancer Group (NCI/CCG) study (1) (Auvinen, Linet et al., 2000) the authors rely on a paper of nigh twenty years ago (2) (Kaune and Gillis, 1981) to argue that no electric field measurements were taken "because of the technical difficulty of relating electric field measurements to subjects’ actual electrical field exposures". Other studies during the 1980s reached similar conclusions. For example, one major study (3) (Savitz, Wachtel et al., 1988) reported spot measurements but abandoned electric field measurements on the basis that their instruments did not permit them meaningfully to address the issue of long term exposure. Moreover the instruments of choice often used in such epidemiological studies (Emdex, Electric Field Measurements Inc.) do not normally include electric field probes. There are however a number of good quality electric field probes available for researfh from perhaps a dozen manufacturers, but these are never provided by EPRI or approved by funding agents. Improvements in instrumentation and dosimetry have made it possible for some years now to capture ambient electric field strength data effectively, and to characterise residential sources of ELF electric fields in dosage terms (4) (Gandhi, 1998). Many studies of low level electric field effects on cells in culture (e.g. (5) Lyle, Ayotte, et al., 1983; (6) Lyle, Schecter et al., 1988; (7) Liburdy et al., 1992) and on living organisms (e.g. (8) Bawin and Adey, 1976; (9) Blackman et al., 1985; (10) Kalmijn, 1982; (11) Morris, Kimball et al., 1989; (12) de Pomerai, Daniells et al., 2000), have yielded valuable insight into the issue. These collectively suggest that measurements of ELF electric fields for short time periods ("spot measurements") in room centres, the method used by Savitz, Wachtel et al, and also (in the case of magnetic fields) by the authors of the NCI/CCG study, poorly characterise the fields around the bedplace itself, which is arguably a more pertinent location. This is becasue electric field strengths can vary tenfold locally within one room as a result of appliance location and internal wiring: in room centres they are likely to be in the range 1-10 V/m but close to appliances and cables may increase to 200-300 V/m (13) (NRPB, 1992, p 19), against which background only bedplace measurements are most likely to reflect long term exposures fairly well. By incorporating data from other rooms it is easy to dilute the exposure profile. In reviewing the residential childhood cancer studies from an electric field viewpoint Auvinen, Linet et al., cite Savitz, Wachtel et al, 1988 and (14) Seversen, Stevens et al., 1988 as failing to find "that electric field exposure was associated with leukaemia either in children or adults". This must be regarded as a shallow analysis, since the data from Savitz, Wachtel et al. 1988 did not include other than spot electric field measurements in room centres, and admitted that their study did not address long term exposure to electric fields, whilst Seversen, Stevens et al., 1988 reported no electric field measurements at all, having confined measurements solely to magnetic fields. Auvinen, Linet et al., further claim that since the commencement of the NCI/CCG study "only one study among four evaluating the relation has reported an association between measured electric fields and the risk of childhood leukaemia". In this regard they cite the (15) London, Thomas et al., 1991 study from Los Angeles County, the same Savitz, Wachtel et al., 1988 study - a mistake, surely, since it was published prior to the NCI/CCG commencement?- , a Canadian study by (16) McBride, Gallagher et al., 1999, and finally that from our own UK laboratory (17) (Coghill, Steward et al., 1996, though it is unfortunately mis-titled in their references). Ignoring the Savitz, Wachtel et al., 1988 study for the reasons already given above, a closer examination of the remainder reveals that only in our own study was the ambient ELF electric field measured for other than short term periods. London, Thomas et al. 1991 only conducted spot measurements of electric fields, using 71 matched pairs, - only about a third out of a total of the 232 matched pairs serving as the basic study population -, and taking measurements "placed as close as possible to the centre of each room and outdoor play area". By contrast the magnetic fields were measured "over at least 24 hours at the place where the child’s bed had been during the etiologic period". Why deliberately use room centre measurements for electric fields but bedplace measurements for the magnetic fields, unless it was the intention to disguise the strength of the E-field association? In that study moreover etiologic period effectively covered the time between birth and up to one year prior to diagnosis whereas the measurements were collected up to two years after diagnosis. Their electric field results only served to show that the mean spot measurements in the centre of bedrooms were very similar in case and control locations, with very large standard deviations (7.48 V/m +/- 9.84 and 7.98 V/m +/-12.3 respectively). When self-reported electric blanket use was evaluated, however, - a situation where very high electric fields are likely, since unlike the magnetic component the electric field is present even when the appliance is not under load, the authors reported a 7-fold odds ratio, though only seven cases were available. The McBride, Gallagher et al., 1999 study was equally inadequate to reflect bedplace ambient electric fields. Though the authors recognised elsewhere (18) (Deadman, Armstrong et al., 1999) that measurements at night provide the best surrogate for total exposure, their protocol for electric field measurement was to use portable electric field dosimeters over a 48 hour period. Therefore during only a small proportion of this period was exposure from the bedplace location collected, and in consequence, since only arithmetic means were reported, bedplace field influence was again diluted. Our own study of 56 UK ALL cases aged 0-14 years and matched controls measured the vertical component of ambient ELF electric and the rms magnetic fields at the bedplace every 30 seconds for a 12 hours period between 2000hrs and 0800hrs the next day. This accordingly provided the most comprehensive assessment yet published of bedplace exposure to ELF electric fields. The results showed a strong dose-response relationship, significantly elevated fields for cases (13.9 V/m) compared with controls (7.3 V/m), and an odds ratio of 4.7 for exposures above 20 V/m. The original concerns expressed in the former Soviet Union over ill health from exposure among high voltage switchyard workers were associated with electric fields, not magnetic (19) (Asanova & Rakov, 1966). A literature review of electric field studies in 1981 (20) (Bridges and Preache, 1981) concluded that the design of studies extant at that time was not adequate for a proper health risk assessment. The former USSR regarded the electric field as the active parameter, however, on the grounds that artificial alternating magnetic components were not dissimilar to those from naturally-occurring semi-static terrestrial levels, whereas alternating electric fields were completely novel in our evolutionary experience (21) (Shandala et al., 1988). Similar levels of electric field exposure among adults have subsequently also been associated with over-representation of leukaemia in a large occupational joint French and Canadian utilities study (22) (Theriault, Goldberg et al., 1994), and other recent UK studies also indict the electric rather than the magnetic component (23) (Fews, Henshaw et al., 1998). The reporting of ELF electric field exposure data collected during the large UK Childhood Cancer study, seriously overdue, made the same classic errors of collection: 3 minute spot measurements and 48 hrs bedplace data, thereby diluting to one third the child's actual bedplace exposure. Was this obvious blunder also done deliberately?. Until their data is re-analysed to consider only the actual likely bedplace exposure, the view that chronic exposure to ELF electric fields may contribute to childhood leukaemia cannot be dismissed. There is an enormous literature implicating weak (non-thermal) electric fields in biological processes and the effects on those processes. Our laboratory has recently shown that even very small ELF and RF/MW electric fields from cellphone handsets on standby can significantly disturb the protective effect of a human donor’s endogenous electric fields on peripheral blood lymphocyte viability (24) (Coghill and Galonja-Coghill, 2000). Fuller investigation of an electric field aetiology for other immune-related disorders generally, and for childhood leukaemias in particular, may go some way towards identifying plausible athermal mechanisms of interaction. The power utilities, the military, the telecommunications and the electrical engineering industries all have vested interests in not doing so. REFERENCES Coghill RW , Steward J, Philips A. Extra Low Frequency electric and magnetic fields in the bedplace of children diagnosed with leukaemia: a case control study. Eur J Cancer Prev 1996; 5: 153-158 etc. |
