American Journal of Epidemiology Vol. 153, No. 2 : 202-205
Copyright © 2001 by The Johns Hopkins University School of Hygiene and Public Health
LETTERS TO THE EDITOR |
RE: "CANCER INCIDENCE NEAR RADIO AND TELEVISION TRANSMITTERS IN GREAT BRITAIN. I. SUTTON COLDFIELD TRANSMITTER; II. ALL HIGH POWER TRANSMITTERS"
West Midlands Cancer Intelligence Unit The Public Health Building The University of Birmingham Birmingham B15 2TT, United Kingdom
Department of Public Health and Epidemiology The Public Health Building The University of Birmingham Birmingham B15 2TT, United Kingdom
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION REFERENCESINTRODUCTION REFERENCES INTRODUCTION REFERENCES |
|---|
Following two earlier studies of hematopoietic and other selected cancers near the Sutton Coldfield television (TV) transmitter in England (Sutton Mast), continuing local concerns prompted a further study in which previous analysis was repeated and more timely cancer data were used. The first study by Dolk et al. (1
) found significant declines in risk for adult leukemia, skin cancer, and bladder cancer with distance from the TV transmitter. A second study by Dolk et al. (2), in which 20 TV transmitters (including Sutton Mast) in the United Kingdom were investigated, was unable to replicate the pattern and the extent of the first study. This research letter reports on whether the results of analysis of adult and childhood leukemia around Sutton Mast in which more recent data were used are consistent with the results from these earlier studies.
Cancer data from the years 1987–1994 were extracted from the West Midlands Cancer Intelligence Unit database (Birmingham, England). The individual cancer sites and cancer groupings (based on International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes) selected for the study followed those used in the earlier national study (2). The total study area was defined as a series of 10 concentric circles centered on Sutton Mast (figure 1). For all cancer sites studied, expected numbers of cases were calculated for each of the 10 bands (persons, male and female adults separately). Expected numbers of all leukemias were also calculated for male and female children (aged 0–14 years). These numbers were based on regional observed cases, stratified by 5-year age bands, sex, and social deprivation. Population data were based on enumeration districts.
|
Stone's statistical methods (3
) were used to examine whether there was a declining risk of leukemia with distance from the source. Stone's test results for all male childhood leukemias (unconditional p value = 0.075, conditional p value = 0.037) suggest some support for a decline in the observed-to-expected (O/E) ratios with distance (table 1). No other cancer group showed a significant decline with distance. Significant p values (Stone's unconditional test) were found for female adult acute myeloid, all leukemia (females and all persons separately), and female hemato-poietic and lymphatic cancers. Over the entire study area (0–10 km), female O/E ratios for these four groups showed a slight increase in risk. Similarly, there was evidence of a slightly elevated risk for leukemias and non-Hodgkin lymphomas combined (sexes combined and separately). There appeared to be no elevated risk in any of the above cancer groups within 2 km of Sutton Mast.
|
If the TV transmitter were a cause of adult leukemia, one would assume both a higher than background incidence in the areas of greatest exposure (i.e., closest to the transmitter) and a decline in risk with distance. This study does not support these assumptions, and no evidence exists of a declining risk of adult leukemia with distance from the transmitter. Evidence of a declining risk in male childhood leukemia with distance is rather weak, and it is difficult to make a causal inference. The original Sutton Coldfield study (1
) found a decreasing risk with distance for adult leukemia and certain subgroups (all acute, acute myeloid, and chronic lymphatic). These conclusions were not replicated.
As an alternative to Stone's test, smoothed O/E ratios by ward for all leukemia (using Bayesian techniques) were calculated to reduce the effects of random fluctuations in cancer rates (4). Intuitive conclusions from inspection of the smoothed maps are that large areas exist in the West Midlands for which there are smoothed ratios above 1.0 for female adult leukemias. The southwest section of our study area appears to be part of one of these areas. Of the 32 wards centered within the study area, 2 were within the highest 5 percent of regional smoothed rates (all female leukemias), and an additional 16 were in the 68–95 percent range. This finding supports other evidence (O/E ratio and Stone's test) of a slightly raised risk in female leukemia over the whole area. More work is planned to examine the validity of using Stone's tests (hypothesis testing) and Bayesian smoothed O/E ratios (ad hoc approach) in epidemiologic studies of this nature.
|
REFERENCES |
|---|
|
TOPINTRODUCTIONREFERENCESINTRODUCTION REFERENCES INTRODUCTION REFERENCES |
|---|
-
Dolk H, Shaddick G, Walls P, et al. Cancer incidence near radio and television transmitters in Great Britain. I. Sutton Coldfield transmitter. Am J Epidemiol 1997;145:1–9.
[Abstract/Free Full Text] -
Dolk H, Elliot P, Shaddick G, et al. Cancer incidence near radio and television transmitters in Great Britain. II. All high power transmitters. Am J Epidemiol 1997;145:10–17.
[Abstract/Free Full Text] - Hills M. Some comments on methods for disease investigation around a point source. In: Elliot P, Cuzick J, English D, et al, eds. Geographical and environmental epidemiology: methods for small area studies. Oxford, United Kingdom: Oxford University Press, 1992:231–7.
- Trinca S. GIS applications for environment and health in Italy. In: Gatrall A, Loyonen M, eds. GIS and health. London, England: Taylor & Francis, 1998:113–24.
Environmental Management and Design Division Lincoln University Canterbury, New Zealand
|
INTRODUCTION |
|---|
|
TOPINTRODUCTIONREFERENCESINTRODUCTION REFERENCES INTRODUCTION REFERENCES |
|---|
A number of studies have shown that the incidence of adult and childhood leukemia is elevated around radio and television broadcast towers in populated areas (1
–6). The most extensive and detailed studies, by Dolk et al. (1, 2), illustrate a major problem with such residential epidemiologic investigations. They reach quite weak conclusions when actually their data provide the strongest published evidence of a causal relation between residential radio frequency (RF)/microwave exposure and cancer, because cancer rates closely follow residential exposure patterns.
Modern broadcast antennas focus their signals toward the horizon, with a succession of side lobes that generally strike the ground within 6 km. The position and strength of the side lobes are a function of the carrier frequency. For very high frequency (VHF) signals (30–300 MHz), the first and strongest peak is inside 1 km. For ultrahigh frequency (UHF) signals (300–3,000 MHz), the strongest peaks are outside 1 km (7, 8). These frequencies produce two characteristic radiation patterns. UHF produces a type A pattern with low exposure close to the tower, rising to an undulating broad peak between 1.5 and 6 km and then declining with distance from the tower. The VHF and mixed VHF/UHF towers have a peak exposure inside 1 km, then a series of less-intense peaks out to about 6 km, and then a continuing decline with distance—a type B pattern.
Because of the radial nature of these patterns, it is important to note that to detect a significant increase in health effects near a tower, three factors need to exist: 1) a high-powered tower with VHF signals; 2) a high population density, because there is only a small land area within 1 km of a tower; and 3) an RF-sensitive health effect. All of the studies cited above show that RF exposure increases the incidence of leukemia. This finding has been confirmed by Milham (9–12) and Szmigielski (13).
In the Dolk et al. studies (1, 2), only one type B cancer pattern existed: adult leukemia at Sutton Coldfield. Sutton Coldfield is the only high-powered site in a high-population area that transmits VHF signals. All other sites and cancer patterns are type A, for adult leukemia and all other reported radial cancer patterns, which includes the typical undulation of the patterns between 1 and 6 km. This makes the chance of confounding vanishingly small and indicates a causal relation between a range of adult cancers and chronic exposure to very low mean-intensity RF radiation at less than 0.1 percent of present standards and the International Commission on Non-Ionizing Radiation Protection guideline.
|
REFERENCES |
|---|
|
TOPINTRODUCTIONREFERENCESINTRODUCTION REFERENCES INTRODUCTION REFERENCES |
|---|
- Dolk H, Shaddick G, Walls P, et al. Cancer incidence near radio and television transmitters in Great Britain. I. Sutton Coldfield transmitter. Am J Epidemiol 1997;145:1–9.
- Dolk H, Elliott P, Shaddick G, et al. Cancer incidence near radio and television transmitters in Great Britain. II. All high power transmitters. Am J Epidemiol 1997;145:10–17.
- Michelozzi P, Ancona C, Fusco D, et al. Risk of leukaemia and residence near a radio transmitter in Italy. Presented at the 1998 ISEE/ISEA Conference, Boston, Massachusetts, August 15–18, 1998. (Paper 354 P). (Abstract published in Epidemiology 1998;9:S111).
- Selvin S, Schulman J, Merrill DW. Distance and risk measures for the analysis of spatial data: a study of childhood cancers. Soc Sci Med 1992;34:769–77.
- Maskarinec G, Cooper J, Swygert L. Investigation of increased incidence in childhood leukemia near radio towers in Hawaii: preliminary observations. J Environ Pathol Toxicol Oncol 1994;13:33–7.[Medline]
- Hocking B, Gordon JR, Grain HL, et al. Cancer incidence and mortality and proximity to TV towers. Med J Aust 1996;165:601–5.[Web of Science][Medline]
- Jones LF. A study of the propagation of wavelengths between three and eight meters. Proc Inst Radio Eng 1933;21:349–86.
- Hammen and Edison Inc. Engineering analysis of radio frequency exposure conditions with addition of digital TV channels. Prepared for Sutra Tower Inc, San Francisco, California, January 3, 1997.
- Milham S Jr. Mortality from leukemia in workers exposed to electric and magnetic fields. N Engl J Med 1982;307:249.[Medline]
- Milham S Jr. Mortality in workers exposed to electromagnetic fields. Environ Health Perspect 1985;62:297–300.[Web of Science][Medline]
- Milham S Jr. Silent keys: leukaemia mortality in amateur radio operators. (Letter). Lancet 1985;l:812.
-
Milham S Jr. Increased mortality in amateur radio operators due to lymphatic and hematopoietic malignancies. Am J Epidemiol 1988;127:50–4.
[Abstract/Free Full Text] - Szmigialski S. Cancer morbidity in subjects occupationally exposed to high frequency (radiofrequency and microwave) electromagnetic radiation. Sci Total Environ 1996;180:9–17.[Medline]
THE FIRST AUTHOR REPLIES
Environmental Epidemiology Unit Department of Public Health and Policy London School of Hygiene and Tropical Medicine London WC1E 7HT, England
|
INTRODUCTION |
|---|
|
TOPINTRODUCTIONREFERENCESINTRODUCTION REFERENCES INTRODUCTION REFERENCES |
|---|
We thank Dr. Cherry (1
) for his comments on our papers (2, 3). In our second study (3), we concluded that lack of replication of the Sutton Coldfield results near other transmitters, notably Crystal Palace, may have occurred because there is in fact no causal link between radio transmission and leukemia or because the model of decline in average exposure with distance is not appropriate for all transmitters. Cherry proposes that the model of decline in exposure with distance is appropriate for FM-only and FM/television (TV) mixed transmitters (such as Sutton Coldfield) but not for TV-only transmitters (such as Crystal Palace). Measure-ments made around Sutton Coldfield, as we described (2), show that field strengths associated with both FM and TV frequencies separately on average decline with distance from the transmitter, although there is considerable variability between different measurement points at any one distance from the transmitter. Cherry does not explain why TV frequencies should show a different pattern of decline in field strength when broadcast alongside FM versus when broadcast alone. He also does not establish how applicable the results in the two studies he cites as evidence (4, 5) are to Crystal Palace in particular. To assess his interpretation, this clarification would be welcome.
Cooper et al. (6) have very usefully carried out further monitoring of cancer statistics in the Sutton Coldfield area in a time period more recent than our study (3). In relation to our main findings, they found a lower excess risk within 2 km of the mast (all leukemia observed/expected ratio = 1.32) with a wide confidence interval (95 percent confidence interval: 0.81, 2.05), no evidence of a decline in risk with distance, and evidence of a generally slightly raised risk of leukemia in the wider area. Given the small numbers and therefore imprecise risk estimates involved, it is difficult to make any unequivocal statements as to whether these results are consistent with or contradict our findings. Further monitoring beyond 1994 is warranted to clarify the situation still further. The findings tend not to confirm the existence of a persistent localized excess near the transmitter of the magnitude we found. Assuming constant exposure to the Sutton Coldfield mast over time and no marked demographic changes, one would expect a continuing localized excess under a causal interpretation of the association between residence near the transmitter and leukemia incidence in the earlier time period.
|
REFERENCES |
|---|
|
TOPINTRODUCTIONREFERENCESINTRODUCTION REFERENCES INTRODUCTION REFERENCES |
|---|
- Cherry N. Re: "Cancer incidence near radio and television transmitters in Great Britain. I. Sutton Coldfield transmitter; II. All high power transmitters." (Letter). Am J Epidemiol 2001;153:204–5.
- Dolk H, Shaddick G, Walls P, et al. Cancer incidence near radio and television transmitters in Great Britain. I. Sutton Coldfield transmitter. Am J Epidemiol 1997;145:1–9.
- Dolk H, Elliott P, Shaddick G, et al. Cancer incidence near radio and television transmitters in Great Britain. II. All high power transmitters. Am J Epidemiol 1997;145:10–17.
- Jones LF. A study of the propagation of wavelengths between three and eight meters. Proc Inst Radio Eng 1933;21:349–86.
- Hammen and Edison Inc. Engineering analysis of radio frequency exposure conditions with addition of digital TV channels. Prepared for Sutra Tower Inc, San Francisco, California, January 3, 1997.
-
Cooper D, Hemmings K, Saunders P. Re: "Cancer incidence near radio and television transmitters in Great Britain. I. Sutton Coldfield transmitter; II. All high power transmitters." (Letter). Am J Epidemiol 2001;153:202–4.
[Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Kheifets and S. Oksuzyan Exposure assessment and other challenges in non-ionizing radiation studies of childhood leukaemia Radiat Prot Dosimetry, December 1, 2008; 132(2): 139 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Schuz, J. Philipp, H. Merzenich, S. Schmiedel, and H. Bruggemeyer RE: "RADIO-FREQUENCY RADIATION EXPOSURE FROM AM RADIO TRANSMITTERS AND CHILDHOOD LEUKEMIA AND BRAIN CANCER" Am. J. Epidemiol., April 1, 2008; 167(7): 883 - 884. [Full Text] [PDF]
|
|
|
|
|
|
RE: "CANCER INCIDENCE NEAR RADIO AND TELEVISION TRANSMITTERS IN GREAT BRITAIN. I. SUTTON COLDFIELD TRANSMITTER; II. ALL HIGH POWER TRANSMITTERS" Am. J. Epidemiol., November 1, 2007; 166(9): 1107 - 1107. [Full Text] [PDF]
|
|
|
|
|
|
M. Ha, H. Im, M. Lee, H. J. Kim, B.-C. Kim, Y.-M. Gimm, and J.-K. Pack Radio-Frequency Radiation Exposure from AM Radio Transmitters and Childhood Leukemia and Brain Cancer Am. J. Epidemiol., August 1, 2007; 166(3): 270 - 279. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Kheifets, M. Repacholi, R. Saunders, and E. van Deventer The Sensitivity of Children to Electromagnetic Fields Pediatrics, August 1, 2005; 116(2): e303 - e313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ahlbom and M. Feychting Electromagnetic radiation: Environmental pollution and health Br. Med. Bull., December 1, 2003; 68(1): 157 - 165. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Michelozzi, A. Capon, U. Kirchmayer, F. Forastiere, A. Biggeri, A. Barca, and C. A. Perucci Adult and Childhood Leukemia near a High-Power Radio Station in Rome, Italy Am. J. Epidemiol., June 15, 2002; 155(12): 1096 - 1103. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Cooper, K. Hemmings, P. Saunders, N. Cherry, and H. Dolk RE: "CANCER INCIDENCE NEAR RADIO AND TELEVISION TRANSMITTERS IN GREAT BRITAIN. I. SUTTON COLDFIELD TRANSMITTER; II. ALL HIGH POWER TRANSMITTERS" Am. J. Epidemiol., January 15, 2001; 153(2): 202 - 205. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||