¹UKCCS, University of Leeds, Institute of Epidemiology, 30 Hyde Terrace, Leeds LS2 9LN, UK

Correspondence: R A Cartwright, E-mail: UKCCS@lrf.leeds.ac.uk

²See end of article for full list of investigators

Received 19 November 2001; Revised 4 March 2002; Accepted 4 March 2002.

Abstract

This paper reports the results of the United Kingdom Childhood Cancer Study relating to risks associated with radon concentrations in participants homes at the time of diagnosis of cancer and for at least 6 months before. Results are given for 2226 case and 3773 control homes. No evidence to support an association between higher radon concentrations and risk of any of the childhood cancers was found. Indeed, evidence of decreasing cancer risks with increasing radon concentrations was observed. Adjustment for deprivation score for area of residence made little difference to this trend and similar patterns were evident in all regions and in all diagnostic groups. The study suggests that control houses had more features, such as double glazing and central heating, leading to higher radon levels than case houses. Further, case houses have features more likely to lead to lower radon levels, e.g. living-rooms above ground level. Consequently the case–control differences could have arisen because of differences between houses associated with deprivation that are not adequately allowed for by the deprivation score.