Study type: Epidemiological study (observational study)

A population-based case-control study of radiofrequency exposure in relation to childhood neoplasm. epidem.

Published in: Sci Total Environ 2012; 435: 472-478

Aim of study (acc. to author)

A population-based case-control study was conducted in Taiwan to investigate the association of radiofrequency exposure from mobile phone base stations and risks of all neoplasms, leukemia, and brain neoplasms in children.

Further details

The annual summarized power (in watt-year (WY)) was calculated for each mobile phone base station annually from 1998 to 2007 from the product of duration of operation (in year) and real emitted power (in watt). The annual power density (WY/km²) in each of the 367 townships in Taiwan was computed as ratio of total annual summarized power from all base stations in a township to the area of that particular township. For each child, the exposure metric was estimated from the averaged annual power density for the five-year period (induction period) prior to the neoplasm diagnosis in the township where the child lived at the time of diagnosis.

Endpoint/type of risk estimation

Type of risk estimation: (odds ratio (OR))



Exposure groups

Group Description
Reference group 1 exposure density < median: < 167.02 WYs/km²
Group 2 exposure density ≥ median: ≥ 167.02 WYs/km²
Reference group 3 exposure density < 1st tertile: 91.65 WYs/km²
Group 4 exposure density ≥ 1st terzile - < 2nd tertile: 91.65 - 392.86 WYs/km²
Group 5 exposure density ≥ 2nd tertile: ≥ 392.86 WYs/km²


Case group

Control group

Study size

Cases Controls
Eligible 2,606 118,170

939 cases of leukemia and 394 cases of brain tumor 78,180; 28,170 and 11,820 children as controls of all neoplasm, leukemia, and brain tumor, respectively

Statistical analysis method: (adjustment: )

Conclusion (acc. to author)

A higher than median averaged annual power density (approximately 168 WYs/km²) was significantly associated with an increased adjusted odds ratio for all neoplasms (OR 1.13; CI 1.01-1.28), but not for leukemia (OR 1.23; CI 0.99-1.52) or brain tumor (OR 1.14, CI 0.83 to 1.55).
The authors conclude that the study does not indicate a significant association of exposure to mobile phone base stations at levels typically encountered in people's everyday environment with risks of brain tumor and leukemia in children. Although risk of overall neoplasm is elevated in children with higher-than-median radiofrequency exposure, the magnitude of risk estimate was small. These results may occur due to several methodological limitations.

Study funded by

Related articles