Mobile phone use and risk of glioma: a case-control study in Korea for 2002-2007

Article information

Environ Health Toxicol. 2015;30.e2015015
Publication date (electronic) : 2015 December 21
doi : https://doi.org/10.5620/eht.e2015015
1Department of Preventive Medicine, Korea University College of Medicine, Seoul, Korea
2Department of Biostatistics, Korea University College of Medicine, seoul, Korea
3Electronics and Telecommunications Research Institute, Daejeon, Korea
4Department of Information and Communication Engineering, Chungbuk National University, Cheongju, Korea
Correspondence: Jae-Wook Choi 73 Inchon-ro, Seongbuk-gu, Seoul 02841, Korea Tel: +82-2-920-6343 Fax: +82-2-920-7220 E-mail: shine@korea.ac.kr
Received 2015 August 3; Accepted 2015 December 11.

Abstract

Objectives

There has been a growing concern about the possible carcinogenic effects of the electromagnetic radiofrequency fields emitted from mobile phones. The purpose of this study was to investigate the association between mobile phone use and the development of gliomas in Korea.

Methods

Our study methods were based on the International Interphone study that aimed to evaluate possible adverse effects of mobile phone use. This study included 285 histologically-confirmed Korean patients 15 to 69 years of age, with gliomas diagnosed between 2002 and 2007 in 9 hospitals. The 285 individually matched controls were healthy individuals that had their medical check-up in the same hospitals. Unconditional logistic regression was used to calculate the adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for use of mobile phones.

Results

For the entire group, no significant relationship was investigated between gliomas and regular use of mobile phones, types of mobile phones, lifetime years of use, monthly service fee, and the other exposure indices. Analyses restricted to self-respondents showed similar results. For ipsilateral users, whose the body side for usual mobile phone use match the location of glioma, the aORs (95% CIs) for lifetime years of use and cumulative hours of use were 1.25 (0.55 to 2.88) and 1.77 (0.32 to 1.84), respectively. However, the contralateral users showed slightly lower risk than ipsilateral users.

Conclusions

Our results do not support the hypothesis that the use of mobile phones increases the risk of glioma; however, we found a non-significant increase in risk among ipsilateral users. These findings suggest further evaluation for glioma risk among long-term mobile phone users.

Introduction

Glioma, the most common primary brain tumor that comprises morphologically distinct cancers such as astrocytoma, ependymoma, and oligodendro glioma [1], is a malignant tumor in the central nervous system with a corresponding high fatality rate. According to the data of the Statistics Korea, 1703 persons with brain or central nervous system tumors accounted for 0.7% of 244177 new cancer cases in 2012 [2]. There are several factors that may have increased the risk of gliomas, including exposure to radiation, genetic drift, and electromagnetic field [3], but the exact causes of gliomas are yet to be determined.

From a health-prospective, concerns have been raised regarding microwaves transmitted from the antenna of a mobile phone could cause brain tumors or increase the risk of the development of potential tumors [4-7] albeit its low-power. In particular, exposure to radiofrequency-electromagnetic field (RFEMF) has been receiving attention due to its negative effect on health, amidst the rapid spread of the use of wireless information service systems.

Most epidemiologic research studies conducted so far on the use of mobile phones and gliomas have failed to show any increase in danger caused by the long-term use of mobile phones [8-13].

However, some studies have reported findings on the relationship between gliomas or brain tumors and the use of mobile phones as follows: the odds ratios (OR) and 95% confidence intervals (CI), i.e., the risk of gliomas when using an analog mobile phone is 1.8 (1.1 to 3.2) [14], the OR (95% CI) for brain tumors based on the use of mobile phones in rural areas is 1.4 (0.98 to 2.0) [15] and 1.12 (0.79 to 1.61) [16], the OR (95% CI) for brain tumors based on the use of analog mobile phones is 2.3(1.2 to 4.1) [14], and the OR (95% CI) associated with use of an analog mobile phone and gliomas is 2.1 (1.3 to 3.4) [17].

In this regard, the International Agency for Research on Cancer (IARC), an international organization for cancer research under the World Health Organization (WHO), examined the validity of epidemiologic research in 1997 in order to investigate the association between the generation of electromagnetic waves from mobile phones and cancer. Subsequently, they started a set of international case-control studies on the link between the use of mobile phones and the development of brain tumors (Interphone study), in which a total of 13 countries participated [18].

The results of recent international epidemiologic studies have increased popular interest in possible health problems and gliomas and other brain tumors owing to the use of mobile phones [4,19-23]. However, little has been known as to their potential mechanisms. Furthermore, there is not enough evidence, either epidemiological or experimental, to support whether RF-EMF has detrimental effect on organisms [24]. Nevertheless, electromagnetic waves from mobile phones may have had a comparatively high effect on the nerves and meningeal tissues close to the surface of the head, thus gliomas and meningiomas become our main concerns.

In 2013, the leading cause of death in Korea was cancer: 144.4 in every a hundred thousand people died of cancer, and 2.4 in every a hundred thousandpeople died of brain tumors.

The use of mobile phones is increasing not only in overseas countries but also in Korea. Statistics showed that subscribers to mobile phones in Korea alone were 47944 thousand in 2009, 50767 thousand in 2010, 52507 thousand in 2011, and 53625 thousand in 2012 [25]. Korea is characterized by relatively high levels in terms of mobile penetration, cumulative hours of use, and lifetime years of use.

Given such an abrupt increase in mobile phone usage, it is important in terms of public health to determine whether electromagnetic waves transmitted by mobile phones are harmful and to carry out the epidemiologic study. Although considerable researches on the effect of electromagnetic waves from mobile phones have been carried out in many countries, there have been few studies that examine the possible link between electromagnetic waves and gliomas in Korea.

Therefore, this study aimed to investigate a possible association between mobile phone use and glioma using a case-control design in nine Korean hospitals based on the protocol of Interphone study [26].

Materials and Methods

Study Subjects

In its final analysis, this study examined 285 patients among 897 patients with gliomas, who were recruited from five areas including Seoul (Seoul, Gyeonggi-do, Gyeongsang-do, Jeolla-do, Chungcheong-do, Gangwon-do, and Jeju-do) and were checked at department of neurosurgery in nine hospitals (Korea University Hospitals (Anam, Guro, Ansan), Inje University Sanggye Paik Hospital, Samsung Medical Center in Seoul, Seoul National University Hospital, Asan Medical Center, Korea Cancer Center Hospital, and Hallym University Sacred Heart Hospital) from 2002 to 2007. The other 612 patients were excluded due to refusal of participation, excessive pain, and impossibility of individual matching (Table S1). The patient group consisted of those who were pathologically diagnosed with gliomas aged between 15 years and 69 years (International Classification of Diseases for Oncology-3 codes 9380-9384, 8391-9460, and 9480) (Table 1) [27].

Histologic type of glioma cases

The control group subjected to the final analysis consisted of 285 healthy persons out of a total of 1051 who randomly received health screenings at the same hospitals as the patient group and were individually matched according to the method for selecting patient-control groups as proposed by the IARC Interphone study team, excluding 766 for the reasons of refusal of participation, excessive pain, and insincere responses (Table S1). The nine hospitals that participated in the study reported patient groups of the diseases to our research team within one week after diagnosis. A questionnaire survey was performed in the individual interview to obtain information for general characteristics and potential confounders. The informed consent was obtained from each subject before enrollment and institutional review board of Korea University approved the study.

Mobile Phone Use Information

Information related with mobile phone use was obtained by a self-administered questionnaire. The types of mobile phone use analog, analog+digital, and digital mobile phones), the lifetime years of use before one year from diagnosis (non-user, <48 months, 48-84 months, and >84 months) the cumulative hours of use (non-user, <300 hours, 300-900 hours, and >900 hours) were obtained. The total amount of mobile phone use was calculated with cumulative hours and lifetime years of use. The monthly service fee was divided into < 30, 30-49, 50-80, and > 80 (unit: 103 Korean won, KRW). Moreover, the average daily receiving call and the average daily sending call were divided into ≤2 times, 3-5 times, 6-9 times, and ≥10 times; and the average call duration time was divided into ≤2 minutes, 3-4 minutes, and ≥5 minutes. The types of mobile phones were classified into flip-type, slide-type, and folder-type; and the regions for carrying a mobile phone during travel were divided into bag, neck, shoulder, pants, and hands. Finally, the usual use of a mobile phone has two categories, rural and urban area.

Confounders or Covariates

Independent variables in the analysis of case-control groups for this study include sex, age (20s, 30s, 40s, 50s, and 60s or older), residential region (Seoul, Gyeonggi, Gyeongsang, Jeolla, Chungcheong, Gangwon, and Jeju), marital status (married, unmarried, others), educational achievement (primary school or lower, middle school or lower, high school or lower, and university or above), annual income level ( < 10, 10-29, 30-49, and ≥ 50, unit: million KRW) respondent (patient in person and proxy: spouse or other family member), and the existence of a transmission tower within 300 m from the residential area. The alcohol drinking (non-drinking, one drink or more a month on average for last one year) and smoking (non-smoking, current smoking) were considered.

Statistical Analysis

To compare characteristics between cases and controls, χ2 test or Fisher’s exact test was performed. The unconditional logistic regression adjusted for sex, age, type of respondent, five residential regions, educational achievement, the use of dye, alcohol drinking, the use of computer, and the use of electric blanket was used to estimate the risk of the brain tumor in relation with usage of mobile phone. Adjusted odds ratios (aORs) and 95% CIs were calculated. As for whether tumor location corresponds to the hand and the body region primarily used during mobile phone call, statistical significance was assessed with relative risk (RR) and p-value the method proposed by Inskip et al. [9], (RR=[√OR+1] ÷2). All the statistical analyses of this study were conducted with SPSS version 12.0 (SPSS Inc., Chicago, IL, USA) and the significance level was 0.05.

Results

The general characteristics of the subjects are as shown in Table 2. Among 285 in the cases, 159 (55.8%) were male and 126 (44.2%) were female. The average ages of the cases and the controls were 42.3 (±14.1) and 42.5 (±14.0), respectively. As for the residential region, 182 resided in Seoul and Gyeonggi, accounting for 63.9% of the overall patients. For the marital status, married (73.7%, 2 in the patient group) were more than the unmarried and others. High school graduates were the most (46%, 131 in the patient group) and for the annual income level, 10-29 million KRW was the highest (49.5%, 137 in the patient group). Furthermore, there were significant differences between cases and controls, in residential region, educational achievement, respondent type, the use of dye, alcohol drinking, the use of computer, and the use of electric blanket.

Distribution of selected characteristics by study group

As for the mobile phone non-users, 9 were male and 37 were female, and the average ages of the cases and controls were 47.3 (±16.0) and 50.4 (±16.4), respectively. There was no significant difference between the mobile phone non-user group and the mobile phone user group in residential region, marital status, and the type of respondent (data not shown).

While the aOR (95% CI) for those who used the mobile phone regularly was 1.17 (0.63 to 2.14), the aOR (95% CI) for the self-respondents was found to be 0.94 (0.46 to 1.89) compared with those who seldom or occasionally used the mobile phone. No significant relationship was found in the regular mobile phone use, the type of mobile phone, the lifetime years of use, the monthly average service fee, and the carriage during travel. However, aOR (95% CI) was 1.92 (0.83 to 4.44) in case the self-respondents used analog and digital simultaneously, 1.35 (0.63 to 2.89) in case the model of mobile phone was folder-type, and 1.42 (0.66 to 3.07) for urban residential region, which was found to be higher than 0.50 (0.22 to 1.13) for rural residential region at a non-significant level (Table 3).

Adjusted Odds ratio (aORs) and 95% confidence intervals (CIs) for risk of glioma in relation to mobile phone exposurea

As a result of analyzing the relation between the body side of usual mobile phone use and the location of glioma using the Inskipet et al.’s method [9], the RR was found to be 1.26 (p=0.05) for the overall respondents with the glioma, and 1.43 (p=0.01) for self-respondents (Table 4).

Laterality of tumor with respect to laterality of telephone use among glioma patients with regular use of mobile phonea

The relationship was adjusted for sex, age, residential region, educational achievement, the use of dye, alcohol drinking, the use of computer, and the use of electric blanket. The risks of glioma for different levels of mobile phone use by ipsilateral and contralateral body side were shown in the Table 5 (total respondents) and Table S2 (self-respondents). In the case of ipsilateral users for total respondents, aORs (95% CI) for the lifetime years of use, cumulative hours of use, the average daily frequency of receiving a call, and the average daily frequency of sending a call were 1.25 (0.55 to 2.88), 1.77 (0.32 to 1.84), 1.52 (0.56 to 4.10), and 3.13 (0.83 to 11.31), respectively, which were found to be high at a non-significant level. On the other hand, contralateral users showed slightly lower aORs (95% CI) than ipsilateral users (Table 5).

Risk of glioma for different levels of mobile phone use by ipsilateral and contralateral among total respondentsa

Discussion

After adjusting for sex, age, residential region, and other variables, this study found no significant relationship between gliomas and mobile phone use, i.e., hours since the initial use of a mobile phone, the period of use, the average daily frequency of receiving a call, the average daily frequency of sending a call, and the monthly average service fee. However, some findings, showed that the risk of gliomas increased at a non-significant level with the folder-type mobile phone and the urban region. It also increased with the simultaneous use of analog and digital phones, the lifetime years of use, the cumulative hours of use, the monthly average service fee, and the average daily frequency of sending a call, for the case of ipsilateral users.

It has been reported that data on the deposition of wireless frequency energy resulting from the form of a car phone or a mobile phone, or the mode of carrying a mobile phone during travel, can be used for an anatomic division of tumorigenesis [28]. The results of this research showed an increased risk, although not statistically significant, when the phone was placed in a shirt pocket among the body regions for carrying a mobile phone during travel. The same results were found for self-respondents. Thus, it is deemed that the location of a mobile phone or the body region has no effect on the development of gliomas. In addition, as a result of this study, it was found that the risk had no particular relationship with a mobile phone service company. The aOR (95% CI) for the folder type mobile phones was 1.35 (0.63 to 2.89) among the self-respondents, which seems to reflect its more users than those of the sliding type, and social and economic status and educational achievement.

In addition to the frequency and duration of mobile phone use, factors that can affect the degree of exposure to micro-electromagnetic waves include the distance from a base transceiver station, localized topography, vegetation, the indoor or outdoor use of a mobile phone, a particular mobile-phone model, the position of an antenna, and the relation between the head and a phone [29,30]. It is difficult to divide and explain these variables with respect to the degree of exposure, and it seemed that these factors were unable to artificially change the frequency of mobile phone use and the time of its use. Therefore, in this study, it was actually possible to investigate the frequency of mobile phone use and the length of its use alone. Some studies revealed that the electromagnetic effect is affected by the use of the electric blanket [31,32].

In this study, the ratio of self-respondents in the patient group and the control group was 86.3% (n = 492) higher than that of proxy respondents. This seems to be resulting from the fact that patients with gliomas do not have the symptoms of disease with hearing loss that can shift the position of using the mobile phone onto the other side, unlike those with other brain tumors or acoustic neurinomas. It seems to be also because with neurologically good condition, they remembered things well and were cooperative with the survey questions.

Overall, epidemiologic studies conducted so far on the relationship between mobile phone use and diseases have not found relationship between the use and cancer genesis [6,8,9]. Moreover, studies on Denmark [10] and Sweden [11] reported that there was no significant relationship between brain tumors and electromagnetic waves emitted by a mobile phone. Also, a case-control study (Interphone study) initiated as an international set of case-control studies in 13 countries around the world [26] focusing on mobile phone use and the risk of brain tumors, showed no risk of gliomas and meningiomas associated with mobile phone use [10,11], which agrees with the findings of this study.

On the other hand, a study reported borderline levels of effects on the risk of gliomas and the use of analog cellular phones, with 432 cases of brain tumor and salivary gland cancer diagnosed in Finland in 1996, with five controls per case [17]. In addition, a case-control study published in 2003 (1617 cases) reported the association of analog mobile phone use with brain tumors [14]. Health hazards, such as an increase in standardized mortality ratio [33] pursuant to the increase of the time of usage [29]and an increase of ocular melanoma occurrence [30] were reported additionally [34].

Overall, findings of studies conducted so far show inconsistent results on the link between electromagnetic waves emitted by mobile phones and brain tumors. Such conflicting results can be attributed to ecological error, inaccuracy of exposure evaluation, and failure to control the information on confounding variables.

In the case where the body side of usual mobile phone use agreed with the location of a glioma (ipsilateral use) for all the respondents, the aOR (95% CI) for the lifetime years of use, the cumulative hours of use, the average daily frequency of receiving a call, and the average daily frequency of sending a call were 1.25 (0.55 to 2.88), 1.77 (0.32 to 1.84), 1.52 (0.56 to 4.10), and 3.13 (0.83 to 11.31), respectively. On the contrary, in the case of disagreement, they showed aOR (95% CI) that was slightly lower. The results correspond to findings of other research on acoustic neurinomas [35] and gliomas [14]. However, some studies on gliomas and meningiomas did not show the increase in risk [11]. In addition, a study involving 678 cases and 3553 controls selected from Sweden, Denmark, Norway, Finland, and two regions in the UK found no relationship of risk with duration of mobile phone use, lifetime years of use or number of calls. Nevertheless, it reported that the risk of brain tumor on the same side of the head as mobile phone use was raised for use for 10 years or longer [36].

In the present study, the analysis result showed an RR of 1.43 (p= 0.01). The study conducted by Inskip et al. [9] showed an RR of 0.9 (p= 0.77) in gliomas, a Japanese study an RR of 0.72 (p= 0.001) in acoustic neurinomas [23] and an Israeli study an RR of 1.32 (p= 0.001) in parotid gland tumors [16], respectively. This showed a consistency among different studies to suggest a stronger association in the same laterality.

The limitations of this study were as follows: First, it is not possible to exclude the possibility of recall bias that might be caused by the patients’ untruthful response, avoidance, memory loss, or exaggeration as to some exposure factors of the questionnaire that included the situation in which the patients knew already they belonged to the patent group. Thus, this study excluded subjects aged 70 or older for the seeming difficulty of deriving accurate responses due to their advanced age. Second, it is probable that selection bias worked because persons relatively more interested in mobile phone radiation participated, and responses could differ between the self-respondents and the proxy respondents. Therefore, this study carried out analysis according to the types of respondents. Third, old subjects of this study aged 60 or older rarely use mobile phones, compared with the young. Therefore, the possibility should be considered that a proper number of samples failed to be obtained. Fourth, during the selection of the patient group, given the dead who died of serious conditions, selection bias may have resulted from including in this study only patients who survived the research period. Thus, it is possible that findings of the current study, which included only mild cases, but excluded the lost, have been underestimated.

Despite the many limitations mentioned above, this case-control study, could secure comparability between the two groups as much as possible. The patient group and the control group were selected by the same standard from the same source population, and the same method was applied to the process of obtaining necessary information from the questionnaire survey. That is, variability could be reduced in collecting information on risk factors since the institutions of the nine hospitals that participated in the research used the standardized common protocol. Furthermore, it was possible to explore the dose-response relationship by grasping the frequency and duration of use, using the personal exposure to the mobile phone found from an additional questionnaire survey on. Such were educational achievement, the type of respondent, the use of dye, alcohol drinking, the use of computer, and the use of electric blanket as well as sex, age, and residential region. This study has the advantage of being the first large-scale research ever performed in Korea on the relationship between gliomas and mobile phone radiation.

Future studies with longer time users and to elucidate the biological mechanism are needed.

Acknowledgements

This work was supported by the IT R&D program of MSIP/IITP (B0138-15-1002, study on the EMF exposure control in smart society).

Notes

The authors have no conflicts of interest associated with material presented in this paper.

Supplementary Material

Table S1.

Numbers of subjects contacted, included and not included in the study

eht-30-e2015015-supple1.pdf
Table S2.

Risk of glioma for different levels of mobile phone use by ipsilateral and contralateral among self-respondentsa

eht-30-e2015015-supple2.pdf

References

1. Sukhdeo K, Hambardzumyan D, Rich JN. Glioma development: where did it all go wrong? Cell 2011;146(2):187–188.
2. Korean Statistical Information Service. 24 Carcinoma/gender/age (5 years old) number of cancer patients, incidence (1995-2012) 2015. [cited 2015 Dec 10]. Available from: http://kosis.kr/statHtml/statHtml.do?orgId=117&tblId=DT_117N_A00023&vw_cd=MT_OTITLE&list_id=117_11744&scrId=&seqNo=&lang_mode=ko&obj_var_id=&itm_id=&conn_path=E1 (Korean).
3. Inskip PD, Linet MS, Heineman EF. Etiology of brain tumors in adults. Epidemiol Rev 1995;17(2):382–414.
4. Rothman KJ, Chou CK, Morgan R, Balzano Q, Guy AW, Funch DP, et al. Assessment of cellular telephone and other radio frequency exposure for epidemiologic research. Epidemiology 1996;7(3):291–298.
5. McKinlay A. Possible health effects related to the use of radiotelephones. Radiol Prot Bull 1997;187:9–16.
6. Hardell L, Näsman A, Påhlson A, Hallquist A, Hansson Mild K. Use of cellular telephones and the risk for brain tumours: a case-control study. Int J Oncol 1999;15(1):113–116.
7. Moulder JE, Erdreich LS, Malyapa RS, Merritt J, Pickard WF, Vijayalaxmi . Cell phones and cancer: what is the evidence for a connection? Radiat Res 1999;151(5):513–531.
8. Muscat JE, Malkin MG, Thompson S, Shore RE, Stellman SD, McRee D, et al. Handheld cellular telephone use and risk of brain cancer. JAMA 2000;284(23):3001–3007.
9. Inskip PD, Tarone RE, Hatch EE, Wilcosky TC, Shapiro WR, Selker RG, et al. Cellular-telephone use and brain tumors. N Engl J Med 2001;344(2):79–86.
10. Christensen HC, Schüz J, Kosteljanetz M, Poulsen HS, Boice JD Jr, McLaughlin JK, et al. Cellular telephones and risk for brain tumors: a population-based, incident case-control study. Neurology 2005;64(7):1189–1195.
11. Lönn S, Ahlbom A, Hall P, Feychting M, ; Swedish Interphone Study Group. Long-term mobile phone use and brain tumor risk. Am J Epidemiol 2005;161(6):526–535.
12. Hepworth SJ, Schoemaker MJ, Muir KR, Swerdlow AJ, van Tongeren MJ, McKinney PA. Mobile phone use and risk of glioma in adults: case-control study. BMJ 2006;332(7546):883–887.
13. Takebayashi T, Varsier N, Kikuchi Y, Wake K, Taki M, Watanabe S, et al. Mobile phone use, exposure to radiofrequency electromagnetic field, and brain tumour: a case-control study. Br J Cancer 2008;98(3):652–659.
14. Hardell L, Mild KH, Carlberg M. Further aspects on cellular and cordless telephones and brain tumours. Int J Oncol 2003;22(2):399–407.
15. Hardell L, Carlberg M, Hansson Mild K. Use of cellular telephones and brain tumour risk in urban and rural areas. Occup Environ Med 2005;62(6):390–394.
16. Sadetzki S, Chetrit A, Jarus-Hakak A, Cardis E, Deutch Y, Duvdevani S, et al. Cellular phone use and risk of benign and malignant parotid gland tumors--a nationwide case-control study. Am J Epidemiol 2008;167(4):457–467.
17. Auvinen A, Hietanen M, Luukkonen R, Koskela RS. Brain tumors and salivary gland cancers among cellular telephone users. Epidemiology 2002;13(3):356–359.
18. Cardis E, Richardson L, Deltour I, Armstrong B, Feychting M, Johansen C, et al. The Interphone study: design, epidemiological methods, and description of the study population. Eur J Epidemiol 2007;22(9):647–664.
19. Violanti JM, Marshall JR. Cellular phones and traffic accidents: an epidemiological approach. Accid Anal Prev 1996;28(2):265–270.
20. Repacholi MH. Low-level exposure to radiofrequency electromagnetic fields: health effects and research needs. Bioelectromagnetics 1998;19(1):1–19.
21. Blettner M, Berg G. Are mobile phones harmful? Acta Oncol 2000;39(8):927–930.
22. Rothman KJ. Epidemiological evidence on health risks of cellular telephones. Lancet 2000;356(9244):1837–1840.
23. Takebayashi T, Akiba S, Kikuchi Y, Taki M, Wake K, Watanabe S, et al. Mobile phone use and acoustic neuroma risk in Japan. Occup Environ Med 2006;63(12):802–807.
24. United Nations Environment Programme, ; International Radiation Protection Association, ; World Health Organization. Electromagnetic fields (300 Hz to 300 GHz) Geneva: World Health Organization; 1993. p. 1–13.
25. Korea Communications Commission. Statistic of mobile phone, 2014. [cited 2015 Dec 10]. Available from: http://www.index.go.kr/potal/stts/idxMain/selectPoSttsIdxSearch.do?idx_cd=2755&stts_cd=275502&clas_div=&idx_sys_cd=654&idx_clas_cd=1 (Korean).
26. Cardis E, Kilkenny M. International case-control study of adult brain, head and neck tumours: results of the feasibility study. Radiat Prot Dosimetry 1999;83(1-2):179–183.
27. Kleihues P. Pathology and genetics of tumours of the nervous system Lyon: International Agency Research on Cancer; 1997. p. 314–320.
28. Balzano Q, Garay O, Manning TJ Jr. Electromagnetic energy exposure of simulated users of portable cellular telephones. IEEE Trans Veh Technol 1995;44(3):390–403.
29. Cleveland RF Jr, Athey TW. Specific absorption rate (SAR) in models of the human head exposed to hand-held UHF portable radios. Bioelectromagnetics 1989;10(2):173–186.
30. Balzano Q. Exposure metrics for rf epidemiology: cellular phone handsets. Radiat Prot Dosimetry 1999;83(1-2):165–169.
31. Li DK, Checkoway H, Mueller BA. Electric blanket use during pregnancy in relation to the risk of congenital urinary tract anomalies among women with a history of subfertility. Epidemiology 1995;6(5):485–489.
32. Robert E, Harris JA, Robert O, Selvin S. Case-control study on maternal residential proximity to high voltage power lines and congenital anomalies in France. Paediatr Perinat Epidemiol 1996;10(1):32–38.
33. Dreyer NA, Loughlin JE, Rothman KJ. Cause-specific mortality in cellular telephone users. JAMA 1999;282(19):1814–1816.
34. Stang A, Anastassiou G, Ahrens W, Bromen K, Bornfeld N, Jöckel KH. The possible role of radiofrequency radiation in the development of uveal melanoma. Epidemiology 2001;12(1):7–12.
35. Lönn S, Ahlbom A, Hall P, Feychting M. Mobile phone use and the risk of acoustic neuroma. Epidemiology 2004;15(6):653–659.
36. Schoemaker MJ, Swerdlow AJ, Ahlbom A, Auvinen A, Blaasaas KG, Cardis E, et al. Mobile phone use and risk of acoustic neuroma: results of the Interphone case-control study in five North European countries. Br J Cancer 2005;93(7):842–848.

Article information Continued

Table 1.

Histologic type of glioma cases

Type ICD-O-3 code n (%)
All cases of glioma 9380-9384, 9391-9460, 9480 285 (100)
Histologic type of tumor 12 (4.2)
Astrocytoma (all grades): grade II 9384, 9400-9421,9424, 9440-9442 56 (19.6)
Anaplastic astrocytoma: grade III 9384, 9400-9421,9424, 9440-9442 32 (11.2)
Glioblastoma and gliosarcoma: grade IV 9384, 9400-9421,9424, 9440-9442 47 (16.5)
Other type 9383, 9391-9394 82 (28.8)
Oligodendroglioma and mixed glioma 9382, 9450-9451 44 (15.4)
Other and unspecified types of glioma 9430-9381,9422, 9423, 9430, 9460, 9480 12 (4.2)

ICD-O-3, International Classification of Diseases for Oncology-3.

Table 2.

Distribution of selected characteristics by study group

Characteristics Cases (n=285) Controls (n=285) p-valuea
Sex Male 159 (55.8) 159 (55.8) 0.86
Female 126 (44.2) 126 (44.2)
Age at interviewb <20 10 (3.5) 8 (2.8) 0.89
20-29 53 (18.6) 48 (16.8)
30-39 68 (23.9) 73 (25.6)
40-49 57 (20.0) 65 (22.8)
50-59 55 (19.3) 55 (19.3)
≥60 42 (14.7) 36 (12.6)
Areac Seoul-Gyeonggi 182 (63.9) 183 (64.2) <0.01
Gyeongsang 43 (15.1) 20 (7.0)
Jeolla 20 (7.0) 34 (11.9)
Chungcheong 29 (10.2) 45 (15.8)
Gangwon-Jeju 11 (3.9) 3 (1.1)
Marital status Married 210 (73.7) 205 (71.9) 0.26
Unmarried 72 (25.3) 70 (24.6)
Othersd 3 (1.1) 10 (3.5)
Education ≤ Primary school 27 (9.5) 27 (9.7) <0.01
Middle school 29 (10.2) 25 (9.0)
High school 131 (46.0) 90 (32.5)
≥ University 98 (34.4) 135 (48.7)
Self-reported annual income (million KRW) <10 44 (15.9) 40 (14.2) 0.41
10-29 137 (49.5) 125 (44.3)
30-49 68 (24.5) 81 (28.7)
≥50 28 (10.1) 36 (12.8)
Respondentse Self 219 (76.8) 273 (95.8) <0.01
Proxy 66 (23.2) 12 (4.2)
Hair coloring No 145 (51.4) 118 (41.5) 0.01
Yes 137 (48.6) 116 (58.5)
Alcohol drinking No 156 (54.7) 128 (45.3) 0.02
Yes 129 (45.3) 155 (54.8)
Cigarette smoking No 229 (80.6) 243 (85.2) 0.14
Yes 55 (19.3) 42 (14.7)
Computer use No 115 (40.4) 91 (32.0) 0.03
Yes 170 (59.6) 193 (68.0)
Watching TV No 16 (5.6) 13 (4.6) 0.57
Yes 269 (94.4) 271 (95.4)
Radio listening No 179 (62.8) 173 (60.9) 0.64
Yes 106 (37.2) 111 (39.1)
Electro-blanket use No 171 (60.0) 202 (71.1) <0.01
Yes 114 (40.0) 82 (28.9)
Transmission towerf No 219 (77.4) 208 (73.2) 0.25
Yes 64 (22.6) 76 (26.8)
a

p-value tested by the Fisher’s exact test.

b

The age of patients at the time of the interview was nearly identical to the age at the time of the diagnosis of the tumor (for patients with tumors) and the age at the time of hospital admission.

c

The area category comprises Seoul Metropolitan City and all the provinces of South Korea.

d

The others category comprises the widowed and the divorced.

e

The proxy respondent is the patient’s spouse or other family member.

f

If the transmission tower is located within less than 300 meters from a patient’s residence, it is considered yes.

Table 3.

Adjusted Odds ratio (aORs) and 95% confidence intervals (CIs) for risk of glioma in relation to mobile phone exposurea

Variable and level of exposure Total respondents
aOR (95% CI)b Self-respondents
aOR (95% CI)c
Cases (n=285) Controls (n=285) Cases (n=219) Controls (n=273)
Use of mobile phone Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
User 239 244 1.17 (0.63, 2.14) 191 235 0.94 (0.46, 1.89)
Type of mobile phone use Non-usera 46 41 1.00 (reference) 25 38 1.00 (reference)
Analogue 22 15 1.83 (0.63, 5.26) 12 15 1.51 (0.45, 5.03)
Analogue+digital 132 119 1.89 (0.96, 3.81) 114 113 1.92 (0.83, 4.44)
Digital 83 109 0.83 (0.43, 1.60) 63 106 0.61 (0.28, 1.33)
Lifetime years of use (mo) Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
<48 49 43 1.28 (0.62, 2.64) 37 41 0.94 (0.42, 2.13)
48-84 88 92 1.27 (0.63, 2.56) 76 89 1.01 (0.45, 2.23)
>84 100 108 1.04 (0.52, 2.09) 76 104 0.90 (0.40, 2.02)
Cumulative hours of use (hr)d Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
<300 97 79 1.25 (0.64, 2.45) 73 77 0.99 (0.46, 2.12)
300-900 70 68 1.59 (0.72, 3.21) 61 67 1.17 (0.53, 2.57)
>900 70 96 0.64 (0.30, 1.34) 55 90 0.62 (0.27, 1.43)
Monthly service fee (103 Korean won) Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
<30 73 57 1.48 (0.73, 3.02) 53 55 1.09 (0.45, 2.47)
30-49 96 111 1.11 (0.57, 2.16) 77 107 0.92 (0.45, 1.98)
50-80 47 55 1.10 (0.52, 2.29) 42 53 0.99 (0.42, 2.29)
>80 22 17 1.12 (0.42, 2.98) 18 16 0.81 (0.28, 2.38)
Average daily receiving call Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
≤2 51 46 1.40 (0.68, 2.88) 43 44 0.95 (0.42, 2.15)
3-5 80 97 1.16 (0.58, 2.31) 64 95 1.00 (0.45, 2.23)
6-9 46 51 1.95 (0.45, 1.99) 38 47 0.89 (0.39, 2.03)
≥10 61 49 1.41 (0.64, 3.09) 46 48 1.20 (0.49, 2.90)
Average daily sending call Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
≤2 73 72 1.44 (0.72, 2.86) 61 69 1.08 (0.49, 2.38)
3-5 82 100 0.97 (0.49, 1.90) 65 96 0.76 (0.34, 1.67)
6-9 30 37 1.15 (0.52, 2.56) 25 36 1.09 (0.45, 2.66)
≥10 53 34 1.65 (0.73, 3.76) 40 33 1.29 (0.51,3.27)
Average duration time (min) Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
≤2 113 116 1.18 (0.62, 2.24) 85 110 0.93 (0.44, 1.96)
3-4 80 81 1.31 (0.65, 2.63) 67 79 1.14 (0.51,2.54)
≥5 45 46 1.00 (0.45, 2.24) 39 45 0.81 (0.33, 1.99)
Shape Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
Flip 19 26 1.51 (0.59, 3.85) 15 26 1.19 (0.41,3.42)
Folder 187 139 1.72 (0.87, 3.38) 148 131 1.35 (0.63, 2.89)
Sliding 30 76 0.55 (0.26, 1.19) 25 75 0.42 (0.10, 1.02)
Carriage Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
In bag 75 74 1.24 (0.61,2.48) 58 72 1.00 (0.45, 2.24)
Hung by neck 12 10 0.76 (0.21, 2.77) 7 7 0.82 (0.20, 3.38)
In shirt 56 51 1.48 (0.68, 3.21) 43 49 1.26 (0.52, 3.06)
In pants 87 95 0.96 (0.47, 1.95) 74 93 0.75 (0.34, 1.66)
At hand 9 14 0.74 (0.24, 2.29) 9 14 0.65 (0.18, 2.25)
Proportion urban/rural at first use Non-usera 46 41 1.00 (reference) 28 38 1.00 (reference)
Urban 174 154 1.66 (0.86, 3.22) 139 147 1.42 (0.66, 3.07)
Rural 65 90 0.63 (0.31,1.30) 52 88 0.50 (0.22, 1.13)
a

Reference category is never or non-regular use of any type of mobile phone.

b

aORs and 95% CIs were derived from unconditional logistic regression for 1:1-matched pairs, with results adjusted for area, education, respondent type, hair coloring, alcohol drinking, computer use and electro-blanket use.

c

aORs and 95% CIs were derived from unconditional logistic regression for 1:1-matched pairs, with results adjusted for area, education, hair coloring, alcohol drinking, computer use and electro-blanket use.

d

For cumulative number and duration of calls category cut-off points were median and 75th percentile.

Table 4.

Laterality of tumor with respect to laterality of telephone use among glioma patients with regular use of mobile phonea

Tumor type Tumor site Total respondents
RRb p-value Self-respondents
RRb p-valuec
Total
Laterality of mobile phone use
Right Left Total Right Left Total
Glioma Right 37 15 52 1.26 0.05 32 13 45 1.43 0.01
Left 22 21 43 14 20 34
Total 59 36 95 46 33 79

RR, relative risk; OR, odds ratio.

a

Patients with tumors whose tumor or telephone use was not exclusively attributed to one side or the other were excluded from the analysis. Laterality was examined by the method proposed by Inskip et al. [9].

b

The RR of a brain tumor associated with mobile phone use was estimated as [(√OR+1)÷2], where OR denotes the unadjusted OR.

c

p-value tested by the Fisher’s exact test.

Table 5.

Risk of glioma for different levels of mobile phone use by ipsilateral and contralateral among total respondentsa

Variable and level of exposure Ipsilateral
aOR (95% CI)b p for trend Contralateral
aOR (95% CI)b p for trend
Cases (n=104) Controls (n=93) Cases (n=83) Controls (n=85)
Use of mobile phones Non-usera 46 41 1.00 (reference) 0.98 46 41 1.00 (reference) 0.35
User 58 52 0.95 (0.50, 1.83) 37 44 0.90 (0.43, 1.89)
Type of mobile phone use Non-usera 46 41 1.00 (reference) 0.96 46 41 1.00 (reference) 0.14
Analogue 2 3 0.42 (0.55, 3.34) 6 2 3.26 (0.52, 20.3)
Analogue+digital 34 29 1.13 (0.52, 2.45) 19 22 0.97 (0.40, 2.33)
Digital 21 20 0.74 (0.32, 1.68) 12 20 0.64 (0.24, 1.70)
Lifetime years of use (mo) Non-usera 46 41 1.00 (reference) 0.98 46 41 1.00 (reference) 0.29
<48 8 4 1.25 (0.29, 5.32) 6 7 0.75 (0.17, 3.17)
48-84 15 22 0.61 (0.25, 1.44) 13 12 1.29 (0.48, 3.46)
>84 34 26 1.25 (0.55, 2.88) 18 25 0.72 (0.29, 1.78)
Cumulative hours of use (hr)c Non-usera 46 41 1.00 (reference) 0.94 46 41 1.00 (reference) 0.25
<300 14 14 0.96 (0.37, 2.47) 14 13 1.20 (0.43, 3.29)
300-900 21 19 1.04 (0.45, 2.40) 9 12 1.09 (0.36, 3.28)
>900 22 19 1.77 (0.32, 1.84) 14 19 0.63 (0.24, 1.65)
Monthly service fee (103 KRW) Non-usera 46 41 1.00 (reference) 0.40 46 41 1.00 (reference) 0.37
<30 15 16 0.86 (0.35, 2.08) 14 13 1.12 (0.39, 3.21)
30-49 29 26 1.04 (0.48, 2.25) 13 16 1.05 (0.40, 2.74)
50-80 8 7 0.72 (0.20, 2.61) 6 12 0.46 (0.12, 1.65)
>80 6 1 4.37 (0.45, 41.9) 4 2 2.26 (0.33, 15.5)
Average daily receiving call Non-usera 46 41 1.00 (reference) 0.43 46 41 1.00 (reference) 0.26
≤2 9 11 0.83 (0.29, 2.41) 9 8 1.05 (0.32, 3.45)
3-5 19 21 0.81 (0.35, 1.91) 12 17 0.97 (0.37, 2.56)
6-9 12 10 0.85 (0.30, 2.42) 6 6 1.18 (0.30, 4.62)
≥10 18 10 1.52 (0.56, 4.10) 9 13 0.57 (0.18, 1.80)
Average daily sending call Non-usera 46 41 1.00 (reference) 0.34 46 41 1.00 (reference) 0.40
≤2 11 14 0.73 (0.28, 1.91) 13 11 1.16 (0.41,3.23)
3-5 26 23 0.97 (0.43, 2.14) 9 19 0.61 (0.22, 1.70)
6-9 6 11 0.44 (0.12, 1.56) 6 6 1.38 (0.36, 5.32)
≥10 15 4 3.13 (0.83, 11.3) 8 8 0.75 (0.21,2.72)
Average duration time (min) Non-usera 46 41 1.00 (reference) 0.88 46 41 1.00 (reference) 0.25
≤2 26 27 2.50 (0.57, 10.9) 18 24 2.65 (0.39, 17.8)
3-4 21 18 1.03 (0.44, 2.42) 12 11 0.84 (0.28, 2.49)
≥5 11 7 0.94 (0.28, 3.09) 6 9 0.65 (0.18, 2.26)

aOR, adjusted odds ratio; odds ratio; CI, confidence interval; KRW, Korean won.

a

Reference category is never or non-regular use of any type of mobile phone.

b

aORs (95% CIs) were derived from unconditional logistic regression for 1:1-matched pairs, with results adjusted for area, education, respondent type, hair coloring, alcohol drinking, computer use and electro-blanket use.