Residential radon exposure and lung cancer : risk assessment based on epidemiologic data

Abstract: Inhalation of decay-products from radon in dwellings is the main source of exposure to ionising radiation for the population of many countries. With an average concentration in dwellings of about 100 Bqm-3 indoor radon levels in Sweden are in the upper range of the average levels observed in other countries. Epidemiologic studies of miners had provided strong evidence of increased lung cancer risk with radon exposure. However, the results from epidemiologic studies of residential radon and lung cancer had shown only weak evidence, and were uninformative in respect of the interaction between radon exposure and tobacco smoking. Additional investigation was warranted to evaluate the risk in the actual settings and levels at which radon exposure occur for the general population. A nationwide case-control study was performed in Sweden, including 1360 subjects with lung cancer diagnosed between 1980 and 1984. For comparison, 2847 subjects were included as controls. Radon was measured in 8992 dwellings occupied by the study subjects during two years or more since 1947. Information on the smoking habits of subjects and other risk factors for lung cancer was obtained from questionnaires. The risk of lung cancer increased in relation to the average radon concentration to which subjects had been exposed, with a percentage increase in risk of 0.10 per 100 Bqm-3 (95 percent confidence interval, 0.01 to 0.22). There was evidence of interaction between tobacco smoking and radon exposure, with individual risks that multiply rather than add to each other. In adjusted risk analyses using empirical data about the variability of radon concentration estimates, the percentage increase in risk was nearly doubled when imprecision in exposure estimates was accounted for. To assess the reliability of group-level analyses, parallel analyses were performed for individual- level and group-level data. No excess lung cancer risk was observed when data was grouped on county-level, in contrast to analyses based on data for individuals. Complementary analyses pointed to a lack of robustness, in county-level analyses, to determinants of lung cancer risk potentially unbalanced across counties. To reduce uncertainty about the radon-related excess relative risk of lung cancer among never- smokers, the database for never-smokers in five Swedish case-controls studies were completed with comprehensive radon measurements. With additional never-smokers from the previous nationwide case-control study, a total of 436 never-smoking lung cancer cases diagnosed in Sweden between 1980 and 1995 were included, as well as 1649 never-smoking controls. Overall, a percentage increase in risk of 10% per 100 Bqm-3 average radon concentration was estimated, which is similar to the summary effect estimate for all subjects in the main residential radon studies to date. The data suggested that, among never-smo ers, residential radon exposure could be more harmful for those exposed to environmental tobacco smoke. A recent glass-based measurement method that could provide more precise estimates for past radon concentrations was applied to glass objects belonging to 110 lung cancer case subjects and 231 control subjects among never-smokers. The relative risks of lung cancer in relation to glass-based average radon concentration estimates were consistently about twice the size of those obtained among these subjects when using air-based average radon concentration estimates. This suggests that glass-based estimates may be more precise than air-based estimates, allowing more accurate risk estimation of the radon-related lung cancer risk, but further validation is needed. In conclusion, residential radon appears as an important cause of lung cancer for the general population. The percentage increase in risk appear similar in smokers and non smokers, corresponding to radon-related risks. of lung cancer that are of an order of magnitude higher among smokers compared to non smokers. As most lung cancer cases occur among smokers, prevention of the largest part of the radon-related excess lung cancer cases is amenable through removal of either smoking or radon.