Exposure to radon has been identified as a risk factor for lung cancer in uranium miners, but evidence of adverse health effects due to indoor radon exposure is inconsistent. Ecological studies have suggested a correlation between indoor radon levels and leukaemia incidence. We evaluated the risk associated with indoor residential radon exposure within a larger interview-based case-control study of risk factors for childhood acute myeloid leukaemia (AML). A total of 173 cases and 254 controls met the eligibility criteria, and information was collected through telephone interviews with parents and analysis of alpha-track radon detectors placed in the home for a period of 1 year. No association was observed between radon exposure and risk of AML, with adjusted odds ratios of 1.2 (95% confidence interval (CI) 0.7-1.8) for 37-100 Bq m(-3) and 1.1 (95% CI 0.6-2.0) for > 100 Bq m(-3) compared with < 37 Bq m(-3). Although there was an inverse association between radon level and AML risk among children < 2 years at diagnosis, among children > or = 2 years, AML risk was increased among those with higher radon exposure. The observed association after age 2 is most likely due to chance. Overall, there was no association between residential radon and risk of childhood AML.