Revision 1
March 19, 1998
by Radiation, Science, and Health, Inc.,
Edited by J. Muckerheide

1.2.6.3
Radon

1.2.6.3.3
Case-Control Studies

Professor Emeritus, and Member of the UN Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), of the Central Laboratory for Radiological Protection, Dr. Zbigniew Jaworowski states (1995b) that: In China, a meticulous study...demonstrated at a 95% confidence level that women in high-level radon houses had an 80% lower lung cancer risk than those living in low-level radon houses. This result is opposite to the no-threshold principle estimate, which claims that the lung cancer risk in high-radon houses should be 80% higher than the normal risk. Similarly, in one region of Japan with an average indoor level of 35 Bq/m3, the lung cancer incidence was 51% of that in a low-level radon region (11 Bq/m3) and the mortality caused by all types of cancer was 37% lower. Similar results showing a lack of positive correlation between lung cancer and indoor radon levels were reported from Canada, Sweden, Denmark, Finland, France, and Great Britain. Despite the evidence from these studies the US EPA recommends remedial action when indoor radon concentrations reach 150 Bq/m3. The EPA considers that remedial action at any level down to 70 Bq/m3 would be cost-effective, even for the cost of reducing the level from 150 to 70 Bq/m3 at approximately $2 million per life hypothetically saved.

Dr. Auvinen and colleagues report in the Journal of the National Cancer Institute (1996) that: In a case-control study in Finland, 1,973 lung cancer cases and 2,885 controls, the odds ratio (OR) of lung cancer for indoor radon exposure was 1.01 after adjusting cigarette smoking. In the analyses stratified by age, sex, smoking status, or histologic type of lung cancer, no statistically significant indications of increased risk of lung cancer related to indoor radon concentration were observed for any of the subgroups. The results indicate no increased risk of lung cancer from indoor radon exposure.

 Case-control studies require precise knowledge of actual individual doses to produce adequate results, to limit the variation in the statistics of small numbers. Knowing the actual dose eliminates the statistical uncertainty in this primary parameter.

 In the case of radon, radon measurements in air in an area do not correlate well to the lung dose to individuals that frequent the area. The correlation is likely to be worse for individual doses in different areas that measure the same radon level. Therefore, doses in radon case-control studies have substantial statistical variation, instead of definitive doses, and radon case-control studies are therefore generally indeterminate. A few studies, under constrained conditions, have adequate radon exposure data to reflect actual doses.

Without knowing doses precisely, only a large statistical database can relate variations in doses to the average values with small error bands that are necessary to produce a strong, repeatable, correlation. Therefore, contrary to the premise that case-control studies are more accurate than ecological studies, for radon this is not true, unless the relationship of the measured radon concentration in air to the dose to each individual can be known with relative precision.

An example of this exception is the case-control study of women in Shenyang China reported below. Because the lifestyle conditions for the women are relatively constrained and stable, and very similar, in typical small residences, over an extended time period, with measured radon data for 1 year in both the kitchen and bedroom, they are more likely to correlate radon in air to lung dose vs other case-control studies. The lung cancer cases correlate with the lower radon exposures, consistent with the more substantial case-control and ecological studies, and contradicting BEIR-IV and BEIR VI predictions.]