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

1.2.6.3
Radon

1.2.6.3.1
Ecological Studies

Professor and Chairman Emeritus Dr. T.D. Luckey, Dept. of Biochemistry of the U. Missouri-Columbia School of Medicine concurs (1994) with Cohen's   conclusions on the health effects of home radon: "There is a strong negative correlation between the radon in homes and lung cancer mortality in males, p<0.001. About 90% of the population of the United States resides in the 1730 counties represented. Cohen obtained comparable data with both males and females. His results were comparable with corrections for smoking."

Professor Luckey also reports (1995) that: "The optimum chronic radon (with progeny) concentration for the reduction of lung cancer mortality appears to be greater than 8 pCi/L. As a public health issue, BEIR IV states that there are 350 lung cancer mortalities per million person-WLM (working level month). Radiation from one WL is equivalent to 100 pCi radon with progeny. Their model suggests home radon would cause 700 lung cancer deaths per million persons at 100 pCi per liter of radon and progeny and 70 per 10 pCi/L. In contrast, Cohen's data indicate that one million people would have 250 fewer lung cancer deaths at 5 pCi/L radon (with progeny) than at 0.5 pCi/L."

Professor Emeritus Dr. Sohei Kondo discusses (Kondo 1993, Section 4.2.1) Cohen's earlier work and confounding factors: Cohen tested the linear no-threshold hypothesis on data covering 411 US counties vs. age-adjusted mortality rates for lung cancer in 1950-69 for white females and males. Then, women were not heavy cigarette smokers, spent a large fraction of their lives at home and seldom migrated from one place to another. Ecological studies are susceptible to confounding. Cohen carried out multiple regression analyses of potential confounding factors. Cigarette smoking is the most important cause of lung cancer. When cigarette sales by state were applied, the negative slope for dependence on radon levels was essentially unchanged. Residents of Misasa [Japan], an urban area where there are radon spas, showed significantly lower mortality from cancers at all sites than residents of the suburbs of Misasa."

Dr. Sadao Hattori, Vice President of CRIEPI reports (1994) that: Cancers in the people of Misasa villages, with high radon levels in drinking water, were compared adjacent villages and all Japan. The negative correlation result was significant.

Professor Emeritus Myron Pollycove, MD, of Laboratory Medicine and Radiology, UCSF, reports (1994) on radon and lung cancer that: BEIR IV predicts that lifetime mortality risk of lung cancer is increased by 10.8% per pCi/L. However, no epidemiologic evidence supports BEIR IV. To the contrary, studies in the US, Sweden, Finland, and China with radon up to 12 pCi/L, as well as in areas below the average radon concentration of 1 pCi/L, have demonstrated a negative correlation of lung cancer with radon. Cohen's 1992 study addresses 54 socio-economic variables (SEV) as confounding factors, and all criticisms of these studies, with excellent statistical power. An unrecognized confounding factor requires: (1) strong correlation with lung cancer, comparable to cigarette smoking, but still unrecognized; (2) strong correlation, of opposite sign, with radon levels; (3) not strongly correlated with any of the 54 SEV; and (4) it must apply to many
geographic areas and be independent of altitude and climate. It must have increased by orders of magnitude, and be more important in males, since the beginning of this century, with effects on females rapidly catching up in recent years. These properties are most difficult to meet singly, while to satisfy the four simultaneously becomes incredible. The more likely explanation of the results is that stimulated biological defense mechanisms more than compensate for the radiation 'insult' and are protective against cancer in a low dose, low dose rate range.

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 also discusses (1995a) home radon and lung cancer: Epidemiological studies of a relation between the radon levels in homes and lung cancer seem to also disagree with the non-threshold principle, and may suggest a hormetic effect. In the US, Cohen's study covering 89% of the population. the people living in houses with radon air concentration higher than average level were found to have a lower mortality from lung cancer (Cohen 1993).

In an abstract, Drs. D. J. Etherington, D. F. Pheby , F. I. Bray also report (1996) on radon and cancer: The incidence of 14 major cancers in Devon and Cornwall were examined vs radon levels. Lung cancer incidence was very similar across all domestic radon categories. Only non-melanoma skin cancers, showed a significant increase in incidence in the high-radon vs low-radon sectors, for both sexes. The remaining 12 cancer sites showed no significant trend with increasing radon concentration. There was no significant difference in corrected survival rates for any cancer site between the low-and high-radon areas.

From the abstract of a study on childhood cancers, Parker and Craft report (1996) that: The effective radiation doses received by children living in high radon areas are similar to those which have been associated with an excess risk of malignant disease elsewhere. However, the only cancer known to be associated with radon is lung cancer, a disease which is not a condition of childhood.

Dr. T. Sobue of the National Cancer Center Institute and his collegues in Japan report (1997) on the Misasa radon area: Misasa town was divided into high and low exposure areas. Compared to Tottori prefecture, the incidences from cancer were generally low, though not statistically significant except that for female in high exposure area (SIR=0.76, 95% CI=0.62-0.91). Stomach cancer in the high exposure area was significantly lower for males (SIR=0.72, 95% CI=0.53-0.96). Caution is needed since individual exposure level was not measured and major confounding factors, such as smoking, could not be controlled in this study.