Abstract
Radon is an established lung carcinogen, but more research is needed to inform whether area-based measures of radon exposure reflect underlying cancer risk. This study, the first of its kind in the UK, aimed to investigate the association between a nationwide indicative radon atlas and lung cancer risk in Northern Ireland, a region in which lung cancer screening programmes are under design. A population-based case-control study was conducted. Primary lung cancer cases (n = 1,687) from the Northern Ireland Cancer Registry diagnosed in 2006 and 2014 were compared with 8,094 non-lung cancer controls from a large cohort. Radon exposure data were extracted from the Indicative Atlas of Radon, which categorises 1 km(2) areas by the percentage of households exceeding 200 Bq/m(3). Exposure data were linked to participants by postcode. Logistic regression models adjusted for age, sex, smoking status, deprivation and PM(2.5) exposure were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for lung cancer. Individuals residing in high radon exposure areas (Atlas Class 6) had a greater than two-fold lung cancer risk (adjusted OR: 2.24, 95% CI: 1.25, 3.92) compared to residents in all other areas. This association was revealed after adjusting for PM(2.5). This study supports the utility of an indicative radon atlas in reflecting underlying population lung cancer risk associated with radon, while highlighting constraints inherent to atlas-based exposure classification and secondary data.