A Combined Analysis of North American Case-Control Studies of Residential Radon and Lung Cancer

Abstract

Cohort studies have consistently shown underground miners exposed to high levels of radon to be at excess risk of lung cancer, and extrapolations based on those results indicate that residential radon may be responsible for nearly 10–15% of all lung cancer deaths per year in the United States. However, case-control studies of residential radon and lung cancer have provided ambiguous evidence of radon lung cancer risks. Regardless, alpha-particle emissions from the short-lived radioactive radon decay products can damage cellular DNA. The possibility that a demonstrated lung carcinogen may be present in large numbers of homes raises a serious public health concern. Thus, a systematic analysis of pooled data from all North American residential radon studies was undertaken to provide a more direct characterization of the public health risk posed by prolonged radon exposure. To evaluate the risk associated with prolonged residential radon exposure, a combined analysis of the primary data from seven large scale case-control studies of residential radon and lung cancer risk was conducted. The combined data set included a total of 4081 cases and 5281 controls, representing the largest aggregation of data on residential radon and lung cancer conducted to date. Residential radon concentrations were determined primarily by α-track detectors placed in the living areas of homes of the study subjects in order to obtain an integrated 1-yr average radon concentration in indoor air. Conditional likelihood regression was used to estimate the excess risk of lung cancer due to residential radon exposure, with adjustment for attained age, sex, study, smoking factors, residential mobility, and completeness of radon measurements. Although the main analyses were based on the combined data set as a whole, we also considered subsets of the data considered to have more accurate radon dosimetry. This included a subset of the data involving 3662 cases and 4966 controls with α-track radon measurements within the exposure time window (ETW) 5–30 yr prior to the index date considered previously by Krewski et al. (2005). Additional restrictions focused on subjects for which a greater proportion of the ETW was covered by measured rather than imputed radon concentrations, and on subjects who occupied at most two residences. The estimated odds ratio (OR) of lung cancer generally increased with radon concentration. The OR trend was consistent with linearity (p = .10), and the excess OR (EOR) was 0.10 per Bq/m3 with 95% confidence limits (−0.01, 0.26). For the subset of the data considered previously by Krewski et al. (2005), the EOR was 0.11 (0.00, 0.28). Further limiting subjects based on our criteria (residential stability and completeness of radon monitoring) expected to improve radon dosimetry led to increased estimates of the EOR. For example, for subjects who had resided in only one or two houses in the 5–30 ETW and who had α-track radon measurements for at least 20 yr of this 25-yr period, the EOR was 0.18 (0.02, 0.43) per 100 Bq/m3. Both estimates are compatible with the EOR of 0.12 (0.02, 0.25) per 100 Bq/m3 predicted by downward extrapolation of the miner data. Collectively, these results provide direct evidence of an association between residential radon and lung cancer risk, a finding predicted by extrapolation of results from occupational studies of radon-exposed underground miners. E. G. Létourneau and J. B. Schoenberg have retired; J. A. Stolwijk holds an emeritus position. We acknowledge the helpful input of the following individuals who served on the International Steering Committee for the North American combined analysis: Ken Chadwick (CEC Radiation Protection Program), Susan Conrath (U.S. Environmental Protection Agency), Sarah Darby (Oxford University), Evan Douple (U.S. National Academy of Sciences), Colin Muirhead (UK National Radiation Protection Board), and Susan Rose (U.S. Department of Energy). Salary support for Drs. Field, Lynch, and Steck was provided in part by grant numbers R01 ES05653 and P30 ES05605 from the National Institute of Environmental Health Sciences, NIH and grant number R01 CA85942 from the National Cancer Institute, NIH. This research was supported by grants from the Canadian Institutes of Health Research (formerly the Medical Research Council of Canada) and the Natural Sciences and Engineering Research Council of Canada to D. Krewski, who currently holds the NSERC/SSHRC/McLaughlin Chair in Population Health Risk Assessment at the University of Ottawa. Financial support for the meetings of the Analysis Team and the Steering Committee was also provided by Health Canada and the U.S. Department of Energy. We are grateful to Dr. Huixia Jiang for assistance with the combined analysis, and to Jackie Monaghan for technical assistance in preparing this report.