Geographic causes

Background: Childhood cancer is a major cause of death in children but the associated risk factors are still largely unknown. Numerous environmental causes, such as background ionising radiation, air pollution, or electromagnetic fields have been suspected and infections in early life may also play a role. Many of these hypotheses imply a spatially heterogeneous distribution of cancer incidence, making them potentially testable with methods of spatial epidemiology. For instance, the putative effects of cosmic ionizing radiation could lead to detectable differences in incidence according to altitude, particularly in a mountainous country like Switzerland. Outbreaks of infections caused by a rapid influx of people into an isolated area could lead to clusters of cancer cases.

Aims: This study aims to investigate 1) whether the incidence of childhood cancer in Switzerland during the period 1985 to present was associated with natural ionizing radiation (cosmic, terrestrial, domestic radon) and traffic related air pollution, and whether there is an interaction between traffic related air pollution and domestic radon 2) whether the incidence of childhood cancer was elevated in communities that experienced rapid population influxes (population mixing) and 3) whether there is evidence for a general tendency of cases to cluster in space and in space-time.

Methods: This study includes the entire Swiss resident population aged <20 years. Data on cancer diagnoses in this population will be obtained from the SCCR and data on the population at risk from the SNC. Outcomes will be any childhood cancer and major diagnostic subgroups such as leukaemia, lymphomas, and tumours of the central nervous system. Among leukaemias, we will separately investigate acute lymphoblastic leukaemia (ALL), and important cytogenetic subtypes of ALL. Dose rates of outdoor cosmic and terrestrial ionising radiation at places of residence will be estimated from existing radiation maps and indoor radon gas concentrations using a previously developed prediction model. Proxy measures for traffic related air pollution will include distance of place of residence to near major roads, to nearest petrol stations or motor vehicle service stations, and predicted outdoor concentrations of specific pollutants.

Significance: The planned work will advance importantly our understanding of the role of environmental risk factors (or the lack thereof) in the aetiology of childhood cancer.

Current status of project: Analyses is ongoing.

Study team: Spycher BD, Kreis C, Kuehni CE, Institute of Social and Preventive Medicine, University of Bern.

Funding: Swiss National Science Foundation (Ambizione, Grant Nr. 147987)

Contact: Ben Spycher (

© ISPM - University of Bern 2019