New studies have provided "significant evidence" of slightly increased cancer risks among people who drink chlorinated water, which includes 75 percent of the nation's population, the Council on Environmental Quality said yesterday.
The chlorine industry did not dispute the finding.
In a report on five independent studies in as many states, the council said the findings showed "increased risks large enough to be of concern, but small enough to be very difficult to separate from . . . risks associated with other environmental factors."
The studies were focused on 11,398 people who died of gastrointestinal or urinary tract cancer in five upstate New York counties, 20 parishes (counties) in Louisiana, 70 Illinois counties outside Chicago and all of Wisconsin and North Carolina. These were matched up with groups similar in age, sex, occupations and other factors, but who died of other causes. The main differentiating factor between the two groups was that more in the first turned out to have drunk chlorinated water.
The report, prepared for CEQ by Science Research Systems Inc., a Ruston, La., consulting firm, stressed that chlorine itself is not the likely culprit. Instead, it is the organic compounds like chloroform that form when chlorine combines with natural and manmade organic material in the water.
"Chlorine almost singlehandedly relegated water-borne diseases to the bottom of the list of causes of death in this country," noted CEQ member Robert Harris."Not using chlorine is not the issue. We can use it in a less risky way."
The studies found risks among chlorinated-water drinkers to be from 13 percent to 93 percent higher for rectal cancer and around 53 percent higher for colon and bladder cancer. Subjects in one study drinking chlorinated rural runoff water were up to three times likelier to have died of rectal cancer than the control group.
That, Harris said, is because rural runoff has more natural organics in the form of fertilizer and decaying vegetable matter than urban runoff does, and therefore more organic compounds will form when chlorine is added.
These compounds are called trihalogenated methanes, and large doses of them have been linked to cancer in laboratory animals. "There's an alphabet soup of organic material that results from chlorination, and all the byproducts aren't even identified yet," Harris said.
Robert Mitchell of the Chlorine Institute in New York noted that the findings appeared to reinforce other studies on the effects of chlorine in water. "I guess we can't dispute it at this point," he said.
The industry produces 12 million tons of chlorine each year and sells it for $1.2 billion wholesale, he said, but chlorine for water is only 5 percent of the business. Chlorine's largest use is in making polyvinyl chloride, a supple plastic used in seat covers and elsewhere, and in bleaching paper, he said.
The five studies were conducted beginning in 1978, the same year that the Environmental Protection Agency proposed regulations to control the formation of the methanes in municipal drinking water. The regulations limit the concentrations to one part per billion parts of water and go into effect next November.
Harris said water treatment centers can begin dealing with the problem by waiting until after the water is filtered or organic materials are otherwise allowed to settle out before adding chlorine. Other disinfectants like ozone or chlorine dioxide could be just as effective and cheap as chlorine in killing unwanted bacteria, he said.
Washington-area water is "on the borderline" of having a high natural organic content and could benefit from relatively simple measures to control it before adding chlorine, Harris said, "but it's not to the point where people should rush out and buy filters."
The CEQ report called for new studies of chlorinated water that would consider its relationship to smoking habits, job stress and consumption of artificial sweeteners, coffee and alcohol, which have all been suspected of possible involvement in gastrointestinal and urinary tract cancer. None of these factors are recorded in the death certificates that provided most of the information in the five current studies, it said, and might still be involved.
The odds of getting cancer shown in the five reports "are small by traditional epidemiological standards and could be produced by a moderate degree of confounding which could go undetected," the report said.