Dirty water.

Will the United States repeat Peru's chlorine folly?

Enrique Ghersi's voice shook with emotion as he spoke to a gathering of Latin American, British, and U.S. policy makers in January. Ten thousand Latin Americans have died of cholera in the past five years, according to the Pan American Health Organization, and more than a million have been made sick. Ghersi, a Peruvian congressman, blames "U.S. imperialism."

But this imperialism didn't come at the point of a bayonet. It came from EPA pronouncements on the chlorination of drinking water.

Chlorine is the most effective killer of bacteria in water supplies. While other techniques such as ozonation or using ultraviolet light can kill bacteria at the point of contact, chlorine added to water keeps on killing all the way to the faucet. About three-fourths of all U.S. drinking water is chlorinated, while most of the rest is treated with a combination of chlorine and ammonia. "Chlorination and disinfection of the water supplies are the public health success stories of the century." says Carol Henry, director of the International Life Sciences Risk Science Institute.

But the broad class of chemicals known as chlorines has been under an all-out assault by American environmentalists lately, who accuse them of everything from thinning the ozone layer to causing cancer to reducing sperm counts to shrinking alligators' penises. Greenpeace International, using the slogan "Chlorine kills!" (meaning people, not germs), has demanded the elimination of all man-made chlorine compounds. Such a drastic move, estimates the consulting firm Charles River Associates, would cost the U.S. economy $91 billion a year. Other critics, pointing to Peru, say it would cost many American lives.

The Peruvian government took the environmentalists and the EPA seriously. Peruvian bureaucrats bought the rhetoric wholesale and greatly reduced the chlorine pumped into the country's water supply.

This action set the stage for horror when, Pan American Health Organization officials suspect, a Chinese freighter released its cholera-contaminated bilge water into Lima's harbor. Eventually the bacteria made its way into open wells, which hadn't been chlorinated, and to other fresh water supplies in which chlorine levels had fallen too low to kill the germ.

As is usual with epidemics, the rich have fared best. Peru's rich can afford private water supplies and bottled water. But the poor, says Ghersi, "have been sacrificed for the EPA and the environmentalists."

Undaunted, the EPA is now taking action that may well endanger much of the U.S. water supply. Citing hypothetical cancer risks from chlorination, the agency in 1994 proposed a rule that would require water systems to eliminate the process known as pre-disinfection. According to some officials, this unfunded mandate would cost local governments an additional $4 million a year and might force small water systems to abandon chlorination completely.

The proposed water regulation has some health officials absolutely boiling. "Reducing chlorine levels in public water systems poses an immediate and significant threat to public health," says Douglas Kinard of the South Carolina Department of Public Health. "A conscientious health professional with experience in water-borne disease outbreaks should not support a rule which compromises public health."

Or, as Robert Forbes of the Florida Department of Health and Rehabilitative Services put it, "The reduction or elimination of chlorination of drinking water to reduce the risk of...disease, is analogous to reducing or eliminating air travel to protect people on the ground from being hit by falling aircraft parts!"

The EPA proposal is particularly disturbing given the lack of evidence that chlorine is in fact harmful. In the past two decades, a dozen epidemiological studies have examined whether chlorine might cause cancer because it can result in the formation of chloroform - which itself has been shown to cause tumors in rodent tests - and because it creates other "disinfection compounds." An overview of this research published three years ago speculated that these byproducts might cause 6,500 cases of rectal cancer and 4,200 cases of bladder cancer.

But these studies show very little increased risk, ranging from 10 to 75 percent. (By contrast, smoking increases the risk of lung cancer by 2,000 percent.) Typically such small increases are considered within statistical "noise," meaning there are just too many things that can throw a study off by that much.

Thus, in a response to a study linking breast cancer and abortion, the National Cancer Institute declared, "Relative risks of less than two are considered small and are usually difficult to interpret. Such increases may be due to chance, statistical bias, or effect of confounding factors that are sometimes not evident."

In the case of chlorine studies, the most glaring problem is probably "recall bias." Study participants were interviewed about risk factors years after their exposure. Quick, ask yourself how much water you drank yesterday. Or the day before? Now imagine being asked how much you drank 10 or 15 years ago.

But won't people be just as likely to overestimate as to underestimate their exposure to chlorinated water? Not if they have already heard from somewhere that there may be an association between chlorinated water and cancer. In that case, the natural tendency is to put two and two together and conclude that the reason you have rectal or bladder cancer is probably because you drank an extraordinary amount of chlorinated water some years earlier. That would make your recollection err on the side of drinking larger amounts of water than you did.

Supposedly one of the most damning aspects of the chlorinated water-cancer connection is that to the extent the studies show any correlation at all, it is a "dose-response" correlation, meaning the, more chlorinated water people drank the more likely they were to get cancer. But that's exactly what you'd expect if recall bias were a problem. In any event, University of California, Berkeley biologist Bruce Ames has noted that if there is any risk of cancer from chlorinated water at all, it is one-thirtieth that of a serving of peanut butter.

Such small risks look trivial amid the death tolls in Peru and elsewhere. And Enrique Ghersi hopes the United States can learn from his country's mistake. "We're ahead of the U.S.," Ghersi told me sarcastically. "And have returned to the Middle Ages as a result."

Michael Fumento is REASON's science correspondent, under a grant from the William H. Donner Foundation, and co-author with Michelle Malkin of a monograph on chlorine, published by the Competitive Enterprise Institute in Washington, D.C.