American Journal of Epidemiology Advance Access originally published online on February 28, 2008
American Journal of Epidemiology 2008 167(7):884-885; doi:10.1093/aje/kwn013
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
LETTERS TO THE EDITOR |
FIVE AUTHORS REPLY
1 Department of Preventive Medicine, College of Medicine, Dankook University, Cheonan 330-714, South Korea
2 Department of Occupational Medicine, Hallym University Hospital, Anyang 431-070, South Korea
3 Electronics and Telecommunication Research Institute, Daejon 305-350, South Korea
4 School of Electronics and Computer Engineering, Dankook University, Seoul 140-714, South Korea
5 Department of Radio Sciences and Engineering, College of Engineering, Chungnam National University, Daejon 305-764, South Korea
(e-mail: minaha{at}dku.edu)
We wish to thank Schüz et al. (1) for their interest in our work (2) and raising some important points about the study.
First, we should respond to the authors' question regarding the estimates in table 2 from our article (2), in which the risks for all types of leukemia were lower even though the risks for the subgroups, both lymphocytic and myelocytic leukemia, were higher. Here, we found a technical error regarding the code for the analyses of radio-frequency radiation (RFR) exposure in the leukemia subtypes. We wish to present the corrected estimates that show a significantly higher risk of lymphocytic leukemia in the highest quartile of exposure for peak RFR, not for the total RFR (table 1). Therefore, although the overall conclusion would not be materially changed, that is, a significantly increased risk of lymphocytic leukemia in children for radio-frequency exposure from AM transmitters, the discussion on the roles of the peak or total RFR among the 31 transmitters should be altered.
|
Second, the authors of the letter (1) suggest on the basis of figure 2 (2) that distance is a good proxy for each single transmitter but not necessarily for all transmitters combined. Although the correlation coefficient between the peak RFR among the 31 transmitters and the distance from the nearest one is higher (Spearman's r = –0.94) than that between the total RFR exposures from the 31 transmitters and the distance from the nearest one (Spearman's r = –0.79), the agreements were caused mainly by the agreement at lower levels of RFR exposure. Therefore, the result would be more obviously biased in the highest exposure group or in those exposed at the closest distance.
Third, the authors of the letter (1) suggest that a separate categorization of the highest percentiles rather than just using the quartile of the exposure would have been more appropriate, because the exposure distribution was highly skewed. When we estimated the odds ratio of lymphocytic leukemia using one more category of the upper 95th percentile of the peak RFR exposure (
1,012.07 mV/m; number of cases = 117), we obtained odds ratios of 1.15 (95 percent confidence interval (CI): 0.85, 1.55), 0.92 (95 percent CI: 0.68, 1.24), 1.45 (95 percent CI: 1.06, 2.00), and 1.24 (95 percent CI: 0.81, 1.91) for the second, third, fourth, and fifth categories of the peak RFR, respectively. There was no significantly increasing effect for the separate highest category of exposure, which might not be distinguishable from that for the fourth category.
Finally, the units of distance and field E (and E', the corrected field) were meters (m) and decibels (voltage level) referenced to 1 µV per meter (dBuV/m). The E (and E') is the electric field from a single transmitter, and the total electric field was calculated from 31 transmitters by use of 
(E'i2), where the unit of E'i is changed to V/m by the conversion equation, dBuV/m = 20 x log10(V/m) + 120. The actual measurement of RFR exposure to validate the prediction program was carried out with a total of 70,365 points at 11 transmitter sites by use of automatic data inquiry and processing software that incorporated a spectrum analyzer and global positioning system loaded in a vehicle.
 |
ACKNOWLEDGMENTS |
|---|
Conflict of interest: none declared.
|
References |
|---|
 TOP References |
|---|
- Schüz J, Philipp J, Merzenich H, et al. Re: "Radio-frequency radiation exposure from AM radio transmitters and childhood leukemia and brain cancer." (Letter). Am J Epidemiol (2008) 167:883–4.
[Free Full Text] - Ha M, Im H, Lee M, et al. Radio-frequency radiation exposure from AM radio transmitters and childhood leukemia and brain cancer. Am J Epidemiol (2007) 166:270–9.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Merzenich, S. Schmiedel, S. Bennack, H. Bruggemeyer, J. Philipp, M. Blettner, and J. Schuz Childhood Leukemia in Relation to Radio Frequency Electromagnetic Fields in the Vicinity of TV and Radio Broadcast Transmitters Am. J. Epidemiol., November 15, 2008; 168(10): 1169 - 1178. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||