The health risk of radon.

Since radon is a serious and widely underestimated health threat, we are publishing this special report to coincide with Radon Action Week (October 15-21) in the hope that this will increase awareness of the radon issue.

Introduction

Although radon is the second leading cause of lung cancer in the United States, second only to cigarette smoking, many members of the public are not aware that radon is one of the most serious environmental cancer risks in the U.S. This paper will outline some of the key. information supporting concern for radon cancer risk.

What is radon? Radon is a naturally occurring, colorless, odorless, tasteless radioactive gas. Its chemically unreactive nature allows radon gas to travel freely through rock and soil. Radon is also moderately soluble in water and can be absorbed by water flowing through radon-containing rock or sand. Uranium-238, a naturally occurring element found in rock and soil, decays through a series of steps to radon-222. Radon-222, with a half life of 3.8 days exists long enough to diffuse through dry porous soils or be transported in water for quite a distance before it decays. Radon-222 undergoes further decay in several steps to produce four radioactive, chemically reactive "daughter" isotopes with short half-lives (less than 30 minutes) which can attach to walls, floors, or airborne particles that can be inhaled. Unattached radon "daughters" can also be inhaled and then deposited on lung tissue. The short half-lives of these deposited radon "daughter" isotopes enables them to undergo considerable radioactive decay before they can be cleared by the mucus action of the bronchial tubes. This radiation can kill, damage, or transform the sensitive cells in the lung. Such a transformed cell has the potential of developing into a cancerous cell.

Based on extensive data from epidemiological studies of underground miners, radon has been classified as a known human carcinogen. In contrast to most pollutants, the assessment of human risk is based on human occupational exposure data rather than animal data. That radon causes lung cancer has been well established by the scientific community, including the World Health Organization's International Agency for Research on Cancer (1), the Biological Effects of ionizing Radiation (BEIR IV) Committee of the National Academy of Sciences (2), the International Commission on Radiological Protection (3), the National Council on Radiation Protection and Measurement (4), the Department of Health and Human Services (HHS), and the EPA.

In addition, the Centers for Disease Control and Prevention, the Surgeon General, the American Lung Association, the American Medical Association, and the American Public Health Association have recognized radon as a significant public health problem.

Studies of Occupational Exposure to Radon

The body of research on lung cancer mortality in underground miners exposed to radon and radon progeny is substantial. Approximately 20 epidemiologic studies of lung cancer mortality in underground miners exposed to radon have been conducted. Studies have been conducted in the United States, Canada, Australia, China, and Europe in metal, fluorspar, shale, and uranium mines. All of these studies have shown an increase in lung cancer risk with exposure to radon decay products, in spite of differences in the mines, miner populations, and study methods (5). For example, the excess relative risk per WLM (working level months) estimates were remarkably similar in five of the major studies of underground miners (6). An international reanalysis of the data on miners led by the National Cancer Institute reviewed 11 studies of male underground miners in the U.S., Canada, Europe, Australia, and China (7). A summary of the report, entitled Lung Cancer in Radon-Exposed Miners and Estimation of Risk from Indoor Radon Exposure, confirms the importance of continued study of the effects of radon exposure (8). More than 2,700 lung cancer deaths among 68,000 male underground miners were included. This analysis confirmed many of the findings on radon in the 1988 National Academy of Sciences BEIR IV Report.

Important findings from the miner studies include:

1) Increased lung cancer risk was observed in miners at relatively low cumulative exposures (e.g., 40-70 WLM -- Ontario uranium miners), similar to what could be obtained for a life-time residential exposure at EPA's action level of 4 pCi/L.

2) The findings indicated that the Excess Relative Risk (ERR) of lung cancer was linearly related to cumulative exposure to radon progeny. The ERR was also found to decrease with attained age, with time since exposure, and with time after cessation of exposure.

3) Miners who smoked had increased lung cancer risk suggesting a synergistic relationship between radon exposure and cigarette smoking (i.e., the risk for smokers exposed to radon is greater than the sum of the two separate risks).

4) Non-smoking miners exposed to radon were shown to have an increased lung cancer risk and had an approximately three-fold higher risk than smokers when compared to their expected baseline.

Based on the results of their analyses, the authors of the report estimated that 15,000 annual U.S. lung cancer deaths may be attributed to radon (95% confidence interval of 6,000-36,000). Of the 15,000 radon-related lung cancer deaths, approximately 10,000 were estimated to be in smokers and 5,000 in never-smokers.

Studies of Residential Exposure to Radon

Because there is always some uncertainty in extrapolating the risk of disease from occupational exposures to the general population, residential radon exposure studies have been undertaken to refine the risk estimate for radon exposure in the home. Residential studies conducted to date can be divided into two basic categories, those examining the association between radon and lung cancer based on group characteristics (ecologic studies), and those examining the association based on individual characteristics (case-control studies).

Ecologic studies may compare rates of exposure and disease between groups of people rather than comparing exposed individuals with and without disease, or they may assess changes in disease frequency and exposure frequency over time for one or more groups. Group selection is often based on geographic parameters. Ecologic studies are relatively in-expensive, convenient, and useful for generating hypotheses, but are not recommended for identifying the cause of a disease. They cannot determine the joint distribution of a specific disease (e.g., lung cancer) and the exposure being studied (e.g., radon exposure) within each group. This study design also can not separate the effects of two or more variables (e.g., smoking and mobility) (9). Use of ecologic data to draw conclusions about effects on individuals (e.g., development of lung cancer) may yield causal inferences that suffer from an "ecological fallacy," i.e., association or lack of association between a study factor and disease observed at group level may not hold at the individual level. Study groups being compared may differ in factors other than those considered in the study and, any such factor or factors may, actually, be the underlying cause of the difference in disease states rather than the factors being studied.

Ecologic studies of radon have generally used characteristics of a geographical region, i.e, geology or radon in water, as surrogates for radon exposure and correlated them with lung cancer rates in the same area. Most showed associations between lung cancer incidence or mortality and exposure to radon. Since individual exposures were not considered, results can only suggest that residential exposure to radon increases lung cancer risk (10).

Because of the limitations of ecologic studies, the 1989 International Workshop on Residential Radon Epidemiology specifically recommended against the further use of ecologic studies and endorsed the use of case-control studies for residential radon research (11).

Case-control studies have been used to more directly test the relationship between lung cancer incidence or mortality and residential radon exposure by comparing the exposure of individuals having the disease of interest, or "cases" and negative controls (12). Compared to ecologic studies, case-control studies are useful for testing causal hypotheses for specific diseases, do not require extensive follow-up, and, since information is collected on individual study subjects, have a greater ability to control for potentially confounding variables (e.g., smoking, diet).

Early case-control studies conducted in Sweden, the United States, and Canada generally used housing characteristics as surrogates for radon exposure and some found associations between radon exposure surrogates and increased lung cancer incidence or mortality.

Seven case-control studies of residential radon exposure based on actual residential measurements and lung cancer risk have recently been completed with mixed results. One study showed a statistically significant dose response relationship (13), one study showed a significant trend of increasing lung cancer risk with increasing radon concentration (14), another showed a borderline significant increase in lung cancer risk with estimated radon exposure (15), three studies showed no statistically significant association between radon exposure and lung cancer risk (16,17,18), and one showed no relationship overall but significant trends in sub-analyses (19). All of these studies have problems such as lack of statistical power (usually due to small numbers of cases), potential confounding by other causes that can obscure the radon/lung cancer relationship, limited exposure data based on lack of access to previously occupied homes, and uncertainties in estimating past exposure from current measurements.

Two International Workshops on Residential Radon Epidemiology held in 1989 and 1991 (20) concluded that many residential radon case-control studies in progress have sample sizes too small to allow characterization of the exposure-response relationship between radon and lung cancer risk. A study by Lubin et al. found that this relationship determination could require tens of thousands of lung cancer cases to meet significant statistical levels (21). These studies individually are currently not helpful in refining the assessment of radon risk to the general population. Further, the potential inability of these studies to detect a consistently significantly increased risk does not acquit radon as a cause of lung cancer at indoor levels (22).

Consequently, study investigators are examining the possibility of pooling results from these completed and other ongoing studies as data become available. Separate efforts are underway to formulate guidelines to pool completed North American studies and European studies. It is hoped that this will ultimately lead to a worldwide pooling effort. However, since these individual studies were not designed to be pooled the success of this effort is not guaranteed.

Some recent studies suggest that in the' future it may be possible to identify the cause of a lung tumor by examining changes in a tumor cells' DNA. Currently, it is not possible to identify lung cancers caused by radon exposure as compared to lung cancers caused by cigarette smoking or other substances. At the cellular level, lung tumors look the same although they have different causes. This may have important implications for future epidemiologic research.

Animal Studies

Data on the relationship of exposure and response from human epidemiologic studies are preferable to animal data when predicting risk for human populations. Uncertainties associated with applying a pattern of exposure and response observed in animals to humans are due to obvious differences including genetic makeup, size, life span, metabolism, substance absorption, and excretion. However, experimental animal studies provide additional confirmatory information on the deleterious health effects associated with radon exposure. Animal studies have also provided useful information to support findings that confounders (ex., silica, dust) are not important contributors to the increased lung cancer risk observed in miners. Animal studies have found that for the same dose, lung cancer risk was greater at low dose-rates (i.e., greater risk when the dose is spread out over a long time period rather than a high dose in a short period).

EPA's Assessment of Radon Risk

To estimate the risk of lung cancer from residential radon, EPA has used the best available scientific data. EPA currently uses a risk projection model developed by the National Academy of Science (NAS); this model is published in the BEIR IV report (23). EPA factored into the risk model information on the difference in dose per unit of exposure between mines and homes. In the Comparative Dosimetry of Radon in Mines and Homes the National Academy of Science concluded that the dose per unit exposure for residential exposures is approximately 30% less than that for a mine exposure (24). The amount of background radon was also accounted for.

EPA's calculated radon risk has been derived using the average national residential radon level (1.25 pCi/L) from the National Residential Radon Survey (25). The Agency also completed a detailed analysis of the uncertainty in lung cancer risk estimates based on an approach developed in a 1985 National Institutes of Health report (26). Sources of uncertainty addressed in this analysis include: 1) statistical variability in the miner data, 2) project of lifetime risk, 3) age dependence of risk, 4) extrapolation from mines to homes, 5) influence of mine exposures other than radon, 6) exposure-rate effect, 7) extrapolation to females, and 8) relationship between radon risk and smoking. Using the modified BEIR IV model, EPA estimates that the number of annual U.S. lung cancer deaths due to residential radon exposures is approximately 14,000 with an uncertainty range of 7,000 to 30,000.

Although we know more about radon than about other cancer-causing environmental pollutants, there is some uncertainty associated with the extrapolation of lung cancer risk from occupational radon exposures to general population residential exposures. However, the average lifetime cumulative radon exposures for the general population are less than an order of magnitude below those for which increased risk can be demonstrated in underground miners. Therefore, extrapolation of risk from occupational exposure to residential radon levels is less of an issue for radon risk assessment than for the assessment of risk for most other environmental carcinogens.

EPA has worked closely with the National Academy of Sciences and other members of the scientific community to use the latest scientific research to estimate risk to the general population from indoor radon. A committee of the National Academy of Sciences, funded by EPA, is currently conducting a comprehensive review of the health risks associated with radon based on new information available since publication of the BEIR IV report in 1988. The updated report will evaluate recent occupational and epidemiologic studies and reassess the risk models presented in the BEIR IV report. Because a full understanding of the risks of low level indoor radon is not likely to come from epidemiologic research alone, the committee will also address the available cellular, molecular, and genetic information. The BEIR VI report is expected to be completed by early 1997. This report and the results of other scientific evaluations may serve to refine EPA's estimate of radon-induced lung cancer deaths.

In summary, although radon is the second leading cause of lung cancer in the United States, second only to cigarette smoking, many members of the public are not aware that radon is one of the most serious environmental cancer risks in the U.S. Based on extensive data from epidemiological studies of underground miners, radon has been classified as a known human carcinogen. In contrast to most pollutants, the assessment of human risk from radon is based on human occupational exposure data rather than animal data. That radon causes lung cancer has been well established by the scientific community. We know more about radon than most other cancer-causing environmental carcinogens. While there are some uncertainties involved when estimating radon risk to the public, it is important to recognize that the risk information is based on human data and that the uncertainties have been addressed in the risk assessment. EPA estimates that the number of annual U.S. lung cancer deaths due to residential radon exposures is approximately 14,000 with an uncertainty range of 7,000 to 30,000. The abundant information on radon health risks that supports EPA's risk assessment indicates that recommendations for public action by the federal government and other public health organizations constitute prudent public policy.