EPRI.com > EMF Health Assessment and RF Safety > EPRI EMF Research in the Literature > Abstracts for Recent Studies  

Abstracts for Recent Studies

An integrated job exposure matrix for electrical exposures of utility workers. Bracken TD, Kavet R, Patterson RM, Fordyce TA. Journal of Occupational and Environmental Hygiene 2009;6(8):499–509. Job-exposure matrices are arrays of rows and columns that match various job titles, tasks, and work environments with exposures workers are likely to encounter on the job. At electric power companies, workers may be exposed to magnetic fields, electric fields, perceptible nuisance shocks, and imperceptible contact currents. Workers may also experience electrical injuries. This paper describes a job-exposure matrix that improves upon previous matrices, which focused on magnetic fields, by addressing all of these factors for 22 job categories. The integrated job-exposure matrix indicates that the highest exposures for all factors combined occur in 4 job categories that involve work near electrical equipment: cable splicers, electricians, line workers, and substation operators.

Future Needs of Occupational Epidemiology of Extremely Low Frequency (ELF) Electric and Magnetic Fields (EMF): Review and Recommendations. Kheifets L, Bowman JD, Checkoway H, Feychting M, Harrington M, Kavet R, Marsh G, Mezei G, Renew DC, van Wijngaarden E. Occupational and Environmental Medicine advance online publication, 19 Sep 2008; doi:10.1136/oem.2007.037994. This paper summarizes the proceedings of a 2006 occupational EMF epidemiology workshop sponsored by the UK's Energy Networks Association. The paper reviews the epidemiologic literature on occupational EMF and health, identifies the highest priority research needs, and proposes steps to address remaining uncertainties. The authors conclude that although the existing epidemiologic evidence does not indicate strong or consistent associations between occupational exposure to EMF and adverse health effects, further research is needed. Identifying exposure assessment improvements and research on the neurodegenerative disease amyotrophic lateral sclerosis (ALS, or Lou Gehrig's disease) as the top research priorities, they recommend development of a holistic job-exposure matrix and an international collaborative study of ALS and electrical occupations.

Occupational Electromagnetic Fields and Leukemia and Brain Cancer: An Update to Two Meta-Analyses. Kheifets L, Monroe J, Vergara X, Mezei G, Afifi A. Journal of Occupational and Environmental Medicine 2008;50:677–88. The aim of this work was to use meta-analysis, a statistical method that combines published data from individual epidemiologic studies, to clarify inconsistent and inconclusive study results on occupational EMF exposure and adult brain cancer and leukemia. As the World Health Organization recommended in its 2007 EMF health risk evaluation, the authors incorporated results from new studies into meta-analyses they published in 1995 and 1997. In addition to the previously included studies, the updated meta-analyses include 20 new brain cancer studies and 21 new leukemia studies. Although combining data from the new studies yielded small risk increases (10–13%) for brain cancer and leukemia, combining data from new and previous studies yielded lower risk estimates for both diseases than those reported in the original meta-analyses. In addition, risk for leukemia subtypes was inconsistent in a comparison of the updated meta-analyses with the previous ones, and there was no clear pattern for workplace EMF exposure and risk of either leukemia or brain cancer. The authors concluded that these results do not support the hypothesis that occupational EMF exposure is responsible for the risk increases.

Exposure to 50 Hz Magnetic Field in Apartment Buildings with Built-In Transformer Stations in Hungary. Thuróczy G, Jánossy G, Nagy N, Bakos J, Szabó J, Mezei G. Radiation Protection Dosimetry advance online publication, 30 Jul 2008; doi: 10.1093/rpd/ncn199. Multilevel apartment buildings with built-in electricity transformer rooms are common in many countries. In this study in Hungary, Thuróczy et al. measured magnetic field levels in apartments in 31 buildings with basement or ground-floor transformer rooms. They found that apartments located immediately above transformer rooms had considerably higher power-frequency (50 hertz [Hz] in Europe) magnetic field levels than those farther away. The authors concluded that the location of apartments relative to transformer rooms reliably predicts magnetic field exposures. These results support the idea that in an epidemiologic study, magnetic field exposures in apartments in buildings with transformer rooms could be assessed without access to apartments or contact with residents. Such a study would avoid selection bias, a form of inadvertent error in epidemiologic studies that arises during the process of study participant selection. This measurement study is part of a feasibility assessment for an international study with minimal selection bias to further investigate the reported epidemiologic association between magnetic fields and childhood leukemia.

Assessment of Selection Bias in the Canadian Case-Control Study of Residential Magnetic Field Exposure and Childhood Leukemia. Mezei G, Spinelli JJ, Wong P, Borugian M, McBride ML. American Journal of Epidemiology 2008;167(12):1504–10. Selection bias is a common methodological error that occurs in epidemiologic studies when those selected for study participation who agree to participate differ in ways that affect study results from those who are not selected or do not agree to participate. This assessment evaluates selection bias in a 1999 case-control study of magnetic field exposure and childhood leukemia in Canada (McBride et al.) that found a weak association. In the original study, the investigators assessed exposure using personal and residential magnetic field measurements and wire coding, a less accurate method based on the characteristics of power lines near residences. In the selection bias evaluation, Mezei et al. used wire coding alone because it is the only method available for assessing exposure for nonparticipants. When they included only actual, participating controls in the analyses, they found a moderate increase in the risk of childhood leukemia for children residing near power lines with the highest wire codes; when they included nonparticipating controls as well, the risk was lower. The authors conclude that although these results suggest that some selection bias may be present in the Canadian study, it may not entirely account for the observed risk increase. They also caution that the use of wire coding rather than field measurements to assess exposure limits interpretation of the results.

Recent Advances in Research Relevant to Electric and Magnetic Field Exposure Guidelines. Kavet R, Bailey WH, Bracken TD, Patterson RM. Bioelectromagnetics 2008;29(7):499–526. This review paper summarizes recent scientific advances relevant to the development and implementation of new or revised EMF exposure guidelines. National and international guidelines limit occupational and public exposure to electric fields, magnetic fields, and contact current (current that flows through the body when it is in simultaneous contact with two conductive surfaces carrying different voltages). Exposure limits are set to prevent known nerve stimulation effects, such as annoyance, startle, and pain. Magnetic field exposure limits are based on prevention of the magnetophosphene effect (perception of a flickering light when exposure exceeds a nerve stimulation threshold, that is, the minimum level for an effect). This paper examines nerve stimulation thresholds and the relevance of magnetophosphenes to guideline limit setting. It also covers dose to body tissues from exposure to contact current and dose to tissues and cells from exposure to spark discharges, or microshocks. In addition, the paper discusses assessment of exposure to high electric fields in real-life situations (such as line work on transmission towers), exposure to nonuniform magnetic fields, and exposures in the workplace.

Calculated SAR distributions in a human voxel phantom due to the reflection of electromagnetic fields from a ground plane between 65 MHz and 2 GHz. Findlay RP, Dimbylow PJ. Physics in Medicine and Biology 2008;53:2277–89. National and international guideline-setting organizations specify limits for exposure to radio-frequency (RF) electromagnetic fields from sources such as radio and television broadcast towers and mobile telecommunications antennas. RF fields deposit thermal energy in the bodies of exposed persons; to protect against excessive heating, guidelines specify basic restrictions limiting the rate at which body tissues may absorb RF energy (the specific absorption rate, or SAR). Because the SAR is difficult to measure, guidelines include limits for corresponding maximum permissible exposures (MPEs) for external field levels, which are easier to measure. This paper describes research to more accurately estimate tissue absorption rates and corresponding field levels. Researchers used accurate computer models of the human body called voxel phantoms to investigate RF energy absorption under various exposure conditions. Results show that guideline basic restrictions and MPE limits provide adequate protection.

Residential Magnetic Field Exposure and Childhood Brain Cancer: A Meta-Analysis. Mezei G, Gadallah M, Kheifets L. Epidemiology 2008;19:424–30. Epidemiologic studies investigating the possibility that residential magnetic field exposure might be associated with childhood brain cancer have yielded inconsistent results. To elucidate the reasons for differences in the results and to provide a statistically robust risk estimate, the authors conducted a meta-analysis of 13 studies. (Meta-analysis is a statistical method that combines published data from individual epidemiologic studies. It is often used when individual studies are too small to permit definite conclusions.) The meta-analysis showed no association of childhood brain cancer with residential distances less than 50 meters from power lines or with wire codes (a surrogate for magnetic field exposure based on power line characteristics) or lower levels of calculated or measured magnetic fields. Although there was a suggestion of an association with measured or calculated fields above 0.3 0.4 microtesla (3 4 milligauss), the association was not statistically significant. The authors conclude that a moderate risk increase cannot be excluded with certainty at higher exposure levels.

Indoor Transformer Stations as Predictors of Residential ELF Magnetic Field Exposure. Ilonen K, Markkanen A, Mezei G, Juutilainen J. Bioelectromagnetics 2008;29:213–8. Epidemiologic studies have reported an association between exposure to magnetic fields and childhood leukemia. However, a causal relationship is not the only explanation: major EMF health risk evaluations note that the association could result from another exposure that is present along with magnetic fields or from inadvertent error in the selection of study participants. To further investigate this possibility, EPRI is planning an international study that will evaluate leukemia incidence among children living in apartment buildings with electricity transformer rooms. The study design avoids errors in participant selection through both selection from cancer and population registries and magnetic field exposure assessment that does not require subject participation. In addition, the study will include larger numbers of children with higher exposures (those whose apartments are adjacent to transformer rooms) than previous studies. This paper reports the results of a preliminary magnetic field measurement study in Finland indicating that exposure in apartments can reliably be predicted according to their location with respect to transformers.

Nighttime Exposure to Electromagnetic Fields and Childhood Leukemia: An Extended Pooled Analysis. Schüz J, Svendsen AL, Linet MS, McBride ML, Roman E, Feychting M, et al. American Journal of Epidemiology 2007;166:263–9. This analysis extended a 2000 pooled analysis of nine childhood leukemia studies (Ahlbom et al.) to determine whether nighttime magnetic field measurements more accurately represent actual exposure than 24- or 48-hour measurements. (Pooled analyses combine original data from individual epidemiologic studies to better discern exposure-disease relationships for a larger number of study participants.) The authors reasoned that nighttime bedroom measurements might be more accurate because children would tend to be in their rooms during the entire measurement period. In addition, nighttime exposure could be more biologically relevant owing to the possibility that magnetic fields might suppress normal nocturnal levels of melatonin, a pineal gland hormone that may protect against cancer development. Results showing similar risk estimates for 24- or 48-hour and nighttime magnetic field exposures do not support these hypotheses.

Survey of Residential Extremely-Low-Frequency Magnetic Field Exposure among Children in Taiwan. Li CY, Mezei G, Sung FC, Silva M, Chen PC, Lee PC, et al. Environment International 2007;33:233–8. Several factors complicate interpretation of epidemiologic results indicating an association between magnetic fields above 0.3 0.4 microtesla and childhood leukemia risk. Among these factors are inadvertent error in study participant selection and the possibility that another exposure occurring along with magnetic fields actually increases risk. Another factor is the unreliability of risk estimates in many studies, owing mainly to small numbers of study participants with higher magnetic field exposures. Future studies to clarify the magnetic field childhood leukemia association will be useful only if they include sufficient numbers of children with higher exposures. A study in Taiwan, a densely populated, industrialized country with reportedly higher residential magnetic field levels, is a possibility. In an EPRI-funded survey, about 5 7 percent of 2214 homes in Taiwan with children under age 7 had measured magnetic field levels above 0.3 0.4 microtesla. These results indicate that a greater percentage of children in Taiwan have higher magnetic field exposures than in North America and Europe, where most epidemiologic studies of EMF and childhood leukemia were conducted.

Assessment of Non-Response Bias in a Survey of Residential Magnetic Field Exposure in Taiwan. Li CY, Mezei G, Sung FC, Silva M, Lee PC, Chen PC, et al. Bioelectromagnetics 2007;28:340–8. In this paper, researchers report the results of an assessment of nonresponse bias in the Taiwan residential magnetic field exposure survey described above. Nonresponse bias is a common form of inadvertent error in the selection of epidemiologic study participants that can occur when people identified as potential study subjects cannot or will not respond to requests to participate. Bias occurs if nonrespondents differ from respondents with respect to exposure or disease status. To assess nonresponse bias in the Taiwan survey, the authors conducted a second magnetic field measurement survey among households that had declined participation and compared the results with those of the original survey. The finding that results are similar indicates little nonresponse bias.

Extremely-Low-Frequency Magnetic Field Exposure of Children at Schools near High Voltage Transmission Lines. Li CY, Sung FC, Chen FL, Lee PC, Silva M, Mezei G. The Science of the Total Environment 2007;376:151–9. This magnetic field measurement study in Taiwan compared children attending schools near high-voltage transmission lines (HVTL) with children whose schools were at least 100 meters from HVTL. The study included both 24-hour personal magnetic field exposure monitoring and measurements at selected classrooms and playgrounds located within 30 meters of HVTL. The results indicate that the two groups of children had a similar mean exposure and a similar proportion of 24-hour exposure above 0.4 microtesla. However, a higher percentage of children at schools close to HVTL had mean exposures greater than 0.4 microtesla during school hours. Mean exposures were particularly high (0.7 microtesla) on playgrounds near HVTL.

Magnetic Field Exposure and Prognostic Factors in Childhood Leukemia. Foliart DE, Mezei G, Iriye R, Silva JM, Ebi KL, Kheifets L, et al. Bioelectromagnetics 2007;28:69–71. This analysis of data from a 2006 study of magnetic field exposure and long-term survival among children with leukemia (Foliart et al.) examined the possibility that magnetic field exposure might be associated with unfavorable prognostic factors. White blood cell count, genetic abnormalities, and other prognostic factors for leukemia are used to estimate the chance that a child will recover from the disease and the chance that the disease might recur after treatment. This analysis found no association between exposure to magnetic fields and the presence of unfavorable prognostic factors.

Magnetic Field Exposure and Long-Term Survival among Children with Leukaemia. Foliart DE, Pollock BH, Mezei G, Iriye R, Silva JM, Ebi KL, et al. British Journal of Cancer 2006;94:161–4. In contrast to previous studies investigating the relation between magnetic field exposure and childhood leukemia incidence (the occurrence of new cases), this study examined whether magnetic field exposure influences relapse and survival rates in children who already have leukemia. The authors report that children whose homes had higher measured magnetic fields (above 0.3 microtesla) experienced more complications during the follow-up period after diagnosis, but this finding was not statistically significant. These children also experienced poorer survival; this finding was statistically significant. However, because these results are based on very small numbers of leukemia cases, they are imprecise. The authors note that independent confirmation of the results is needed since the study is the first of its kind.

Socioeconomic Status and Childhood Solid Tumor and Lymphoma Incidence in Canada. Mezei G, Borugian MJ, Spinelli JJ, Wilkins R, Abanto Z, McBride ML. American Journal of Epidemiology 2006 advance online publication, 8 March 2006; doi:10.1093/aje/kwj118. This study follows up a 2005 study (Borugian et al.) in which the same team of researchers used neighborhood income to measure socioeconomic status (SES) among childhood leukemia cases identified from Canadian cancer registries. In the 2005 study, children from the poorest neighborhoods had a modestly decreased risk of acute lymphoid leukemia, the most common form of childhood leukemia, compared to children from the richest neighborhoods. In the new study, the relationship between SES and other types of childhood cancer was examined. A moderately lower risk of carcinomas and renal tumors was observed among the poorest children. Although these results could indicate a relation between SES and these types of cancer, the authors note that they could be due to chance. No consistent relation was observed between SES and various other childhood cancers; this may argue against a causal role for environmental exposures that are strongly linked to SES.

Physical Activity and Magnetic Field Exposure in Pregnancy. Savitz DA, Herring AH, Mezei G, Evenson KR, Terry JW, Jr., Kavet R. Epidemiology 2006;17:222–5. Two 2002 studies by Lee et al. and Li et al. reported that high peak magnetic field exposure (the highest exposure encountered during a day) was associated with increased miscarriage risk. However, previous evidence provides little support for a magnetic field?miscarriage association. In a commentary published along with the 2002 studies, epidemiologist David Savitz suggested that the association might be explained by differences in physical activity between women who had normal pregnancies and women who miscarried: owing to less nausea and vomiting in early pregnancy and more mobility and energy in later pregnancy, women who miscarried would move around more, encountering more sources of high magnetic fields (for example, household appliances, office equipment, and electric power lines). To test this hypothesis, Savitz and his team investigated the relation between physical activity level, measured with an activity meter, and magnetic field exposure among pregnant women. They found that women with higher activity levels were more likely to encounter high peak magnetic fields. These results support Savitz's hypothesis, but more research is needed to address the relation between physical activity and symptoms associated with pregnancy outcomes.

Analyses of Magnetic-Field Peak-Exposure Summary Measures. Mezei G, Bracken TD, Senior R, Kavet R. Journal of Exposure Analysis and Environmental Epidemiology advance online publication, 12 October 2005; doi:10.1038/sj.jea.7500457. To shed light on the magnetic field?miscarriage association reported by Lee et al. and Li et al. in their 2002 studies, this analysis investigated the characteristics of peak magnetic field exposure measures. The analysis examined activity level information and magnetic field exposure data from the Li et al. study and three previous studies that measured personal exposure to residential magnetic fields. The results showed that the magnitude of measured peak magnetic fields depended on the sampling interval set for the exposure meter and that maximum measurement values varied when measurements were repeated. Also, study subjects (both men and women) with higher activity levels had higher peak magnetic field exposures. This analysis lends support to the hypothesis that the association between magnetic fields and miscarriage in the 2002 studies may be due to higher activity levels among women who miscarry.

Selection Bias and its Implications for Case-Control Studies: A Case Study of Magnetic Field Exposure and Childhood Leukemia. Mezei G, Kheifets L. International Journal of Epidemiology advance online publication, 22 November 2005; doi:10.1093/ije/dyi245. EMF health risk evaluation panels have noted that the association between magnetic fields and childhood leukemia observed in epidemiologic case-control studies could at least partly result from selection bias, a form of inadvertent error that may arise during the process of study participant selection. The authors of this case study examined epidemiologic studies of magnetic fields and childhood leukemia to evaluate the potential for selection bias in these studies. They found evidence both for and against selection bias; in many studies, however, reporting of selection processes was inaccurate and incomplete, making evaluation difficult. The authors conclude that better reporting and evaluation are needed, along with new methods for selecting and recruiting controls.

Childhood Leukemia and Socioeconomic Status in Canada. Borugian MJ, Spinelli JJ, Mezei G, Wilkins R, Abanto Z, McBride ML. Epidemiology 2005;16:526–31. Early childhood leukemia studies reported a higher leukemia incidence in children from families with higher socioeconomic status (SES). However, more recent case-control studies of magnetic field exposure and childhood leukemia have reported a higher incidence among children with lower SES. To investigate whether the shift in incidence is real or a result of unintentional error (bias) due to case selection or study participation, researchers used neighborhood income as a measure of SES in a study of childhood leukemia cases identified from population-based Canadian cancer registries. They found that children in the poorest neighborhoods had a moderately lower risk of acute lymphoid leukemia, the most common form of childhood leukemia, than children in the richest neighborhoods. These results suggest that high SES may be a risk factor for childhood leukemia and that inconsistent results in previous studies may stem from differences in case selection or study participation.

The Interaction between ELF Electric Fields and RF Survey Meters: Theory and Experiment. Olsen RG, Yamazaki K. IEEE Transactions on Electromagnetic Compatibility 2005;47:86–96. Radio-frequency (RF) survey meters, used to measure workers’ personal exposure to RF electromagnetic fields, may give erroneous readings in the presence of strong extremely low frequency (ELF) fields. This paper presents theoretical and experimental information that contributes to accurate assessment of electric power company worker exposure near high-voltage transmission towers and distribution facilities hosting RF communications antennas.

Animal Models for the Study of Childhood Leukemia: Considerations for Model Identification and Optimization to Identify Potential Risk Factors. McCormick DL, Kavet R. International Journal of Toxicology 2004;23:149–61. In this paper, authors David McCormick of IIT Research Institute and Rob Kavet of EPRI discuss the technical challenges involved in identifying and optimizing a mouse model suitable for studying the potential role of environmental agents in childhood leukemia development. The paper won the American College of Toxicology President’s Award for the best paper published in 2004 in the International Journal of Toxicology.

Association of Residential Magnetic Fields with Contact Voltage. Kavet R, Zaffanella LE, Pearson RL, Dallapiazza J. Bioelectromagnetics 2004;25:530–6. Grounding of the electrical service in a U.S. home to the home’s water line, as required by the National Electrical Code, results in a voltage between the water line and the earth. This voltage, in turn, drives a voltage between water fixtures and conductive drain pipes that can be a source of contact current exposure to a bathing child touching the water fixtures or water stream. In this study of 191 single-family Denver homes, both voltages were positively associated with spot-measured average residential magnetic fields. These results support the hypothesis that exposure to contact current may be responsible for the association found in epidemiologic studies between magnetic fields and childhood leukemia.