Differences in lung cancer mortality trends from 1986-2011 by radon risk areas in British Columbia, Canada

摘要

Background: Radon gas is a decay product of uranium. Residential exposure is associated with increased lung cancer risk, especially in smokers. Most evidence about the health effects comes from international meta-analyses, but administrative data can help public health authorities to explore local impacts. Aims: To examine and compare temporal trends in lung cancer mortality from 1986-2011 in low moderate, and high radon areas of British Columbia (BC), Canada. To illustrate differences by sex and smoking prevalence. Methods: More than 3,800 residential radon concentrations were assembled from four agencies. Eighty local health areas were classified as having low, moderate, or high radon risk based on the number of homes with concentrations greater than the Health Canada action guidelines of 200 and 600 Bq/m3. Vital statistic records were used to identify deaths due to lung cancer and due to all natural causes. The annual ratio of lung cancer mortality to all natural mortality was plotted over the study period for each radon classification. Analyses were stratified by sex and by the 2008 prevalence of current smokers. Results: The fitted ratio increased from 8% to 10% over the study period in high radon areas, and remained stable at 7.5% in the low and moderate radon areas. The high radon areas showed different patterns when stratified by sex and smoking prevalence. For males the increase was from 9% to 10%, and for females it was from 6% to 11%. For the lower smoking group the ratio increased from 6% to 9% during the first half of the study period, and then stabilized. For the higher smoking group the ratio was stable at 9% for the first half of the study period, and then it increased to 11%. Conclusions: These findings provide evidence that radon is an important risk factor for lung cancer in BC, and they suggest that the burden of disease could be reduced by targeting smokers in high radon areas. The results would be strengthened by replication, and more rigorous methods are required to assess other contributing factors.