BENER Abstract No. 15047. LEUKEMIA FOLLOWING OCCUPATIONAL EXPOSURE TO 60-HZ ELECTRIC AND MAGNETIC FIELDS AMONG ONTARIO ELECTRIC UTILITY WORKERS.
(Eng.) Miller, A. B.; To, T.; Agnew, D. A.; Wall, C.; Green, L. M. [Dept. of Preventive Medicine and Biostatistics, Faculty of Medicine, Univ. of Toronto, 12 Queens Park Crescent West, Ontario, M5S 1A8 Canada (RR/A.B.M., T.T., C.W.); Health and Safety Services, Ontario Hydro, Toronto, Ontario, Canada (D.A.A., L.M.G.);] Am J Epidemiol 144(2):150-160; 1996 (33 Refs).
The authors reported the results of a further analysis of leukemia related to occupational exposure of electric utility workers to 60-Hz electric and magnetic fields in a nested case control study within a study base of active and retired male employees of Ontario Hydro, Ontario, Canada. Active workers were followed from 1973 until 1988 and retired workers known to be active on 1 January 1970, were followed from 1973 through 1988. The cohort was part of a larger study of cancer in employees of Electricite de France, Hydro Quebec, and Ontario Hydro utilities described previously by Theriault et al. (Am J Epidemiol 139:550-572, 1994; BENER Abstract No. 10018). An excess of leukemia was found in the Theriault et al. study which occurred primarily in the Ontario Hydro component, so possible associations of cancer with electric field exposure were evaluated in this cohort and previously reported findings for magnetic fields in the Ontario workers were reevaluated. New male employees who entered the workforce and those who became pensioners from 1973 to 1988 were also included in the study base. This created a cohort of 31,543 male Ontario Hydro employees. The period of observation for cancer detection began after 1 yr of full employment and ended on 31 December 1988, or with the death of the worker, whichever came earlier. Cancer cases were identified by linking the cohort to records of the Ontario Cancer Registry. The major emphasis was on melanoma, all malignant and benign brain tumors, and all hematopoietic malignancies which had previously been associated with EMF exposure and were designated as "a priori cancers." Controls were randomly selected from the files of Ontario Hydro. The selection criteria required controls to have the same year of birth, to be alive in the year of diagnosis of the case, and to have no previous diagnosis of cancer other than non-melanoma skin cancer (basal cell and squamous cell carcinoma). Exposures to electric and magnetic fields and potential occupational confounders (chlorophenoxy acid herbicides, benzene, asbestos, cadmium compounds, sunlight, or ionizing radiation) were estimated through job exposure matrices (JEMs). The JEMs were constructed from job titles and direct magnetic and electric field measurements obtained for a group of 895 workers in representative jobs who had worn Positron monitors for 5 working days. Quantitative estimates of occupational ionizing radiation exposure were made using information contained in the Ontario Radiation Dose Information System. Possible confounding by socioeconomic status was estimated using a 5 category system thought to represent the employee status at the time of hire. Associations between electric and magnetic field exposure and cancer were examined by logistic regression techniques. The final study group consisted of 1,484 cancer cases and 2,179 matched controls. Of the cancer cases, 233 were a prior cancers and most (230) could be matched to 4 controls (2 were matched to 3 controls, and 1 to 2 controls). Pensioners constituted 46.2% of the total study population and 40.6% were still actively working at the time of diagnosis. The remaining subjects had left Ontario Hydro before the end of the observation period (31 December 1988). Cumulative electric and magnetic field exposures were similar for cases and controls. Approximately 63.8% of the cases and 61.0% of the controls, for example, experienced cumulative electric field exposures of 100-499 V/m-yr. None had cumulative electric field exposures of greater than 4,999 V/m-yr. Approximately 53.8% of the cases and 54.1% of the controls had cumulative magnetic field exposures of 2.5-9.9 uT-yr. Only 1 case and 1 control had a cumulative exposure as high as 250-uT/yr. Of the 233 cancers of a priori interest, 140 were hematopoietic malignancies, 35 were brain tumors, and 58 were malignant melanomas. Of the 140 hematopoietic malignancies, 50 were leukemias. Of these, 13 were acute myeloid leukemia (AML), 8 were chronic myeloid leukemia (CML), 19 were chronic
lymphocytic leukemia (CLL), and 10 were unspecified leukemia. Twenty cases were also classified as acute nonlymphoid leukemia (ANLL), a category which included the 13 AML cases and 7 of the unspecified cases. Lung cancer (263 cases) and prostate cancer (244 cases) represented the largest number of malignancies at other sites. Elevated odds ratios (ORs) were observed for all the a priori cancers except malignant and benign brain tumors. When examined by tertiles of cumulative electric field exposure (0-171, 172-344, and 345 or greater V/m-yr), and, after adjusting for the potential confounders, the risk for all leukemia was significantly increased for cumulative exposure to 345 V/m-yr or stronger when compared to the lowest tertile of exposure (adjusted OR = 4.45, 95% confidence interval (CI) 1.01-19.7, 26 cases). The risks for ANLL, AML and CLL in the highest exposure tertile were nonsignificantly increased, ORs = 7.89, 24.53, and 7.18, respectively. There was a suggestion of a dose-response trend for most of the a priori cancer sites, but none of the trends was statistically significant. For other cancer sites, the risks for stomach and lung cancer were nonsignificantly increased. When examined according to tertiles of cumulative magnetic field exposure (0-3.1, 3.2-7, and 7.1 uT-yr or stronger), the crude ORs for all leukemia, ANLL, and AML in the middle and highest exposure tertiles were increased, but decreased after adjustment. The adjusted ORs for malignant and benign brain tumors were increased, but not significantly. For example, in the 7.1 uT-yr or greater exposure group, the crude and adjusted ORs for malignant brain tumors were 1.33 and 2.36 (CI 0.52-10.8, 10 cases), respectively. When cumulative electric and magnetic field exposures were considered together, leukemia risk increased with increasing electric field exposure within each tertile of magnetic field exposure. For example, within the 7.1 uT-yr or greater tertile of magnetic field exposure, the ORs for all leukemia in the 0-171, 172-344, and 345 V/m-yr or greater electric field exposure tertiles were 1.20 (CI 0.11-13.2, 1 case, 11 controls), 7.83 (CI 1.06-58.0, 6 cases, 20 controls), and 11.2 (CI 1.30-97.2, 17 cases, 34 controls), respectively. Within each tertile of electric field exposure there was no evidence of an increasing risk with increasing magnetic field exposure. The combined OR for leukemia for high exposure to electric fields and low exposure to magnetic fields was 4.27 (CI 1.00-18.17); for high exposure to electric fields and medium magnetic field exposure, the OR was 4.79 (CI 1.03-22.22), and for high exposure to both electric and magnetic fields, the OR was 5.53 (CI 0.88-34.63). An analysis by job title showed that there were increased leukemia risk ORs for operators (3.34), power line maintenance workers (1.89), and technical, maintenance, and security workers (2.37), but none were significantly greater than unity. By job site, nonsignificantly elevated leukemia ORs were seen for persons employed in area offices (2.69), hydroelectric (2.33), nuclear (2.91), and thermal (2.05) generating stations, and transformer stations (4.00). This study confirms the previously reported associations between estimated exposures to magnetic fields and certain types of leukemia in the Ontario Hydro component of the Theriault et al. study, but also demonstrates significant new associations between cumulative electric field exposure and all leukemia. There are also suggestions that the risk may primarily relate to ANLL and its component, AML. The findings also suggest that the failure to incorporate job site may have resulted in misclassification in previous studies. Also, prior associations of leukemia with magnetic field exposure may have been driven by simultaneous exposure to electric fields.
BENER Digest Update is a current awareness journal published every March, June, September and December. The Digest summarizes the results of key EMF bioeffects research published in a broad spectrum of professional journals. It is the only print publication that regularly surveys the world's literature on EMF biological effects, including research that directly relates to possible adverse health effects from EMF exposure. In-depth, original abstracts provide detailed experimental results and multidisciplinary interpretation. Full bibliographic references plus author and subject indexes help the reader maintain a current overview of this expanding field. The Digest also features original articles on EMF biological effects. Ongoing research, meeting abstracts, and summaries of popular-press news articles are included as well. Each quarterly issue puts all the most recent EMF research at your fingertips in one concise journal.