Radon (specifically radon-222), a well-documented cause of lung cancer in underground miners, is a contaminant of indoor environments, including homes and schools. This odorless and invisible gas is a naturally occurring decay product of radium-226, the fifth decay product of uranium-238. Uranium-238 and radium-226 are constituents of most soils and rocks, so radon is found in the gas present in soils. It is also released from water in which it has dissolved and from building materials. Radon decays with a half-life of 3.82 days into a series of solid, short-lived radioisotopes that are collectively referred to as radon progeny (formerly radon daughters) or radon decay products. Two of the progeny, polonium-218 and polonium-214, emit particles during decay. When these emissions take place in the lung as inhaled and deposited progeny undergo decay, the cells lining the airways can be damaged in such a way that lung cancer can eventually occur.
Radon has long been known to cause lung cancer (for a review, see National Research Council, 1988). More than 100 years ago, miners of metal ores in Schneeberg, Germany, were found to
develop intrathoracic malignancy, which was shown to be primary cancer of the lung. Early in the twentieth century, high concentrations of radon were measured in the Schneeberg mines and in the nearby mines of Joachimsthal, where underground miners also developed lung cancer. Radon was considered to be a possible cause of the lung cancer, and this was confirmed through epidemiologic studies of miners of uranium and other ores (for a review, see National Research Council, 1988; Samet, 1989).
Many populations of underground miners exposed to radon and its progeny have been shown to be at increased risk of lung cancer. Except at the highest levels of exposure, the lung-cancer risk in these miners is related roughly linearly to exposure. The information available from miners on the combined effect of cigarette smoking and exposure to radon progeny is consistent with synergism between the two carcinogens. Exposure of animals to radon has provided confirming evidence of carcinogenicity, and laboratory systems have been used to understand mechanisms of genetic injury by particles. The newer techniques of molecular and cellular biology are now being applied to -particle carcinogenesis; the initial findings indicate the potential for these techniques to improve understanding.
Research over the last 20 years has shown that radon is a ubiquitous indoor air pollutant, reaching concentrations in some residences as high as were found in mines where excess lung cancers occurred in underground workers. The predominant source of radon in indoor air in homes is the soil beneath the structures, but building materials, water used in the homes, and utility natural gas can also contribute. Radon concentrations are readily measured with passive devices, and early data showed that the distribution of concentrations was approximately lognormal with a mean of about 1.5 picocuries per liter (pCi/L) (Nero et al., 1986). Data gathered from over 4,000 U.S. homes in 1989-1990 showed that the average
concentration is 1.3 pCi/L (U.S. Environmental Protection Agency, 1992). However, the distribution is skewed with a tail extending well beyond the average; some homes have concentrations of hundreds or even thousands of picocuries per liter. Although detailed data are available on samples of homes, far fewer measurements have been made in other locations where people spend time.
Since the recognition that a carcinogen contaminates indoor air particularly in homes where most time is spent-risk assessment has been used to estimate the extent of the hazard as a basis for risk management. Given a lack of direct evidence from epidemiologic studies on indoor radon and lung cancer, risk models for exposures received by the general population (U.S. Environmental Protection Agency, 1992) have been based on extrapolation from higher exposures in studies of underground miners. Extending the findings in the underground miners to the general population entailed various assumptions each with its own uncertainties. Some of these assumptions are:
The risks observed at occupational exposure levels and dose rates can be extended to typically lower indoor exposure levels and dose rates.
The modifying effects of smoking on the risk associated with exposure to radon progeny are similar in miners and the general public.
The risk of lung cancer in radon-exposed miners is not substantially modified by effects of other agents, e.g., dust.
The risks observed in adult male miners can be extended to females and to children.
Either exposure-dose relationships are comparable in miners and the general population or, if there are differences, the differences can be estimated.
In spite of the uncertainties inherent in risk assessments based on the miner data, indoor radon has been identified as an important public-health problem, estimated by EPA to cause between 7,000 and 30,000 lung-cancer deaths a year (U.S. Environmental Protection Agency, 1992). The lung-cancer risk associated with indoor radon can be estimated from epidemiologic studies that directly assess the risk in exposed populations. When the possible risks of exposure to indoor radon were first recognized, several descriptive or ''ecologic" studies (the use of groups-most often defined geographically-rather than individuals as the unit of analysis) were performed; lung-cancer incidence or mortality rates in geographic areas were compared with indexes of radon exposure for inhabitants of these units. However, the evidence from those studies has been inconsistent, and methodologic limitations seriously limit the value of the ecologic design for characterizing the lung-cancer risk associated with indoor radon (Stidley and Samet, 1993).
The case-control design is a more appropriate epidemiologic approach for addressing indoor radon. Most of the many reported case-control studies have had small numbers of subjects or low exposure estimates. Several larger studies are now in progress in the United States and other countries (Samet et al., 1991b). In these studies, past exposures to radon are estimated from present concentrations in current and previous residences of lung-cancer patients and appropriate controls; with information collected on cigarette-smoking and other risk factors, the risk of indoor radon can be assessed and the effects of smoking and other factors can be controlled for. The findings of these studies are limited by uncertainties in the estimation of past exposures and by other methodologic problems (Lubin et al., 1990).
Regardless of the limitations of the data available, risk assessments based on the studies of miners suggest a need for concern
regarding a positive association between the level of indoor radon and the occurrence of lung cancer; therefore, indoor radon should be considered a potential public-health problem. The scope of the exposed population and the large number of lung-cancer cases attributed to radon have provided a strong impetus for population and laboratory-based research designed to provide a more complete understanding of radon carcinogenesis and more accurate risk assessments.