Several new studies of radon, the radioactive gas known to cause cancer and found to be seeping into millions of homes across the country, have uncovered little evidence linking household exposure to disease, raising questions about how much risk radon poses to humans at very low levels.
Researchers say these studies have in most cases failed to show an association between lung cancer and household radon levels at or even slightly higher than the level at which the Environmental Protection Agency recommends taking corrective measures.
The continued lack of evidence of a significant risk from radon at these levels raises questions about how aggressively the nation should try to lower radon levels in homes and buildings. The ultimate cost to meet the EPA standard is estimated at $50 billion or more, critics say.
But the EPA and other scientists contend there is enough indirect evidence that radon is a major cause of cancer at levels found in many homes to take action. The design and limited number of participants in some of the new studies fail to give them sufficient power to assess the true risks, they say.
Radon is an odorless, colorless gas that arises naturally from the ground because of the decay of radioactive elements commonly found in rocks and many types of soil.
In a chain of radioactive decay, uranium produces radium, which gives off radon, which in turn produces radioactive breakdown products that are harmful if inhaled.
Almost all scientists agree that prolonged exposure to high levels of radon is hazardous, but uncertainty remains about the risks of low-level exposure. Worldwide studies of lung cancer in underground miners, backed by animal research, show that long-term exposure to high levels of radon and its breakdown products causes lung cancer and that the effect is heightened by cigarette smoking, researchers say.
Most radon experts concur that studies show that radon gas is a genuine cancer hazard to humans at levels two to four times and above the EPA “action level” for home exposure, which is 4 picocuries per liter of air or higher. A picocurie is one-trillionth of a curie, a unit of radioactivity.
According to EPA surveys, the average radon level in American homes is 1.25 picocuries, and 6 percent of all homes, or about six million, have levels of 4 picocuries or above.
But several long-awaited studies designed to quantify the radon risk in homes have not found a significant cause-effect relationship at typical levels of exposure.
Dr. Jay H. Lubin, an epidemiologist with the National Cancer Institute who has analyzed numerous radon studies, including the newest ones, said all had weaknesses that precluded determining the exact risk of radon in homes.
“Evidence showing an excess risk from indoor exposure to radon remains inconclusive,” he said. “The results of most of these studies is consistent with there being no effect at low levels, but there is still reason to think radon in homes is deleterious.”
The data on miners are decisive in showing that radioactive particles from the decay of radon causes lung cancer, Lubin said, and there is no evidence of a threshold below which any exposure would not increase risk.
However, he said, differences in the environments of underground mines and homes could greatly affect actual exposures to radon, making extrapolating risks from mines to houses difficult.
Based on the miner data, indoor radon exposure is estimated to be responsible for about 10 percent of the 150,000 lung cancer deaths that occur in the United States each year.
In one of the recent studies, researchers at Health Canada’s Health Protection Branch looked at 738 lung cancer cases in Winnipeg and an equal number of people with similar characteristics who had not developed lung cancer. They found no increase in risk attributed to cumulative radon exposure among long-term residents.
The researchers, including Dr. Ernest G. Letourneau and Dr. Daniel Krewski, chose Winnipeg because it had the highest indoor radon levels of 18 Canadian cities surveyed. The mean household exposure for those in the study was 3.5 picocuries per liter and measurements in the basements averaged 4.5 picocuries, said the study, published last month in The American Journal of Epidemiology.
Krewski said a strength of the Canadian study was to try to pinpoint total exposure by measuring radon levels in every home each person in the study had lived in; many other studies have measured radon in only one residence. Although researchers took measurements in 4,448 homes, however, more than a third of the eligible homes could not be tested.
In another study published in the journal Health Physics last March, researchers at the University of Kansas School of Medicine studied hundreds of women who were long-term residents of 20 counties in Iowa with above-average radon levels.
Dr. John S. Neuberger and his colleagues said there was a correlation between radon and lung cancer in counties with high rates of smoking, and decreasing radon-associated cancer rates in areas of less smoking.
Dr. Michael Alavanja and colleagues at the National Cancer Institute studied the lung cancer cases of 600 non-smoking women in Missouri compared with 1,400 healthy women and found no clear association with household radon concentrations.
However, the study, which will soon be published in The Journal of the National Cancer Institute, did find a suggestive-but not statistically significant-trend linking higher home radon levels with one type of tumor.
Over all, it looks as if the data are not showing an effect of radon, Alavanja said in an interview, “but I think there may be something going on at four picocuries and above.”
The study that found the most positive association between lung cancer and radon was published in January in The New England Journal of Medicine. This Swedish study reviewed the cases of 1,360 men and women with lung cancer and compared them with 2,847 healthy people. It found that people exposed to 3.8 to 10.8 picocuries had a cancer risk 30 percent higher than those whose exposure averaged below 1.4 picocuries.
However, some other researchers expressed concerns about the study, noting, for instance, that radon sampling was done in the winter, when levels in closed-up houses are highest, and year-round exposures were extrapolated from these readings.
