Abstract
The aetiology of most childhood cancers remains poorly understood. We conducted a nationwide register-based case-control study to assess the association between residential road traffic and railway noise exposure and risk of childhood cancers.
We identified all cases of first cancers diagnosed in children aged 0-19 years in 1985-2013 from the Danish Cancer Registry (N = 3962) and sampled four individually matched (by sex and date of birth) controls per case (N = 14,790) using the Central Population Register. We estimated time-weighted exposure averages of residential road traffic and railway noise at the most (L
max) and least (L
min) exposed façades from birth to index-date (for additional analysis: in utero period) based on the individual address history for the respective time windows. We fitted conditional logistic regression models to estimate odds ratios (OR) and their 95% confidence intervals (CI).
ORs varied by noise estimate and cancer type, with generally wide CIs mostly including 1.00. We found a tendency of higher ORs with increasing railway and road traffic noise for Hodgkin lymphoma (ORs for railway and road L
min were 1.63 (95% CI 1.00; 2.66) and 1.14 (95% CI 0.87; 1.48) per 10 dB), as well as a tendency of higher ORs with increasing railway noise for non-Hodgkin lymphoma. For embryonal CNS tumours and astrocytoma and other glioma we observed also some weak suggestions of a positive association. Analysing exposure to traffic noise in utero revealed similar patterns to those of the main analyses.
This nationwide study with minimal risk of bias suggests no strong associations between traffic noise and risk of most childhood cancers. We found however some suggestive evidence for a positive association with Hodgkin lymphoma, non-Hodgkin lymphoma and some CNS tumours. Further research is warranted to confirm these associations in other populations and elucidate the underlying biological mechanisms.