Not Entirely Clear
In Virginia, Water Chlorination Is Blamed for Miscarriages. Below the Surface, the Science on the Issue Is Murky

By Christopher Wanjek
Special to The Washington Post
Tuesday, March 12, 2002; Page HE01

Chlorine is a near-perfect disinfectant, most health experts agree, killing deadly bacteria and other tiny
organisms in the water supply cheaply and effectively and saving hundreds of thousands of lives annually
in the United States.

The Centers for Disease Control and Prevention (CDC), in fact, ranks chlorinated water among the
greatest public health achievements of the 20th century, right up there with penicillin and vaccines. Thanks
to chlorine treatment, the chronic outbreaks of cholera and dysentery that plagued crowded American
cities through the 18th and 19th centuries are gone.

Ah, if only chlorine didn't cause cancer and now, maybe, miscarriages. Plaintiffs in a high-profile
Virginia lawsuit claim a byproduct of chlorination has caused some women in their community to have

While the benefits of chlorination are beyond debate, the science regarding its potential harms is

Chlorine reacts with organic material in the water supply dead leaves, vegetation, farm runoff to
produce chemicals called trihalomethanes (THM), some of which cause cancer in laboratory animals
exposed to extremely high doses. Based on years of animal and human studies suggesting, but not
establishing, a causal relationship between THMs and cancer, the Environmental Protection Agency
(EPA) suspects that THMs in tap water cause a small number of bladder cancers in humans. As a result,
the EPA set a limit for THM concentrations in 1979 at 100 parts per billion (ppb). It lowered this to 80
ppb, effective January 2002.

THMs' effect on pregnancy is far less certain, according to the EPA. A billion-dollar class action lawsuit
has been filed in Chesapeake, Va., where residents fear that drinking tap water led to miscarriages in the
1980s and 1990s. So far, animal studies have shown only that high doses of THMs can cause birth
defects, not miscarriages. And in human studies, "the direct evidence is quite weak now," said David
Savitz, who chairs the Epidemiology Department at the University of North Carolina (UNC) School of
Public Health in Chapel Hill.

Savitz, who has studied this issue for over a decade, has recently embarked on the largest study to date
of THMs' possible link to miscarriage. The project, due to be completed in 2005, is being funded by the
American Water Works Association, whose members include major utility companies.

Complicating this research, Savitz said, are the facts that even routine miscarriage rates are hard to
quantify and that estimating direct exposure to THMs is extremely complex.

No Chlorine: A Bad Thing

It is sparklingly clear to health experts that an absence of chlorine in the water is a bad thing. The World
Health Organization estimates that 25,000 children die daily worldwide from illnesses associated with
contaminated drinking water.

Cholera was absent from the Americas for nearly a century until an outbreak occurred in Peru in 1991
and spread to neighboring countries. More than 4,000 people died that year, 6,000 more have died since
then and the threat of cholera lingers, according to the Pan American Health Organization.

"There's no question that a lack of chlorine allowed cholera to spread in Peru," said R. Bradley Sack, a
professor at the Bloomberg School of Public Health at Johns Hopkins University. The outbreak was the
result of a gradual breakdown of Peru's water quality control during the prior decade, he said.

The Science of Miscarriage

Only in developed nations, where cholera and dysentery are virtually nonexistent, have the possibly
deleterious effects of chlorination become a health issue. The trick is finding a balance between the deadly
effects of too little chlorine in the water and the possible long-term effects of too much.

The primary studies showing a relationship between tap water and miscarriages were conducted in
California in the 1990s. The researchers found that, in one of the three counties studied, women who
drank tap water were at least twice as likely to experience miscarriages as those who drank bottled
water. (There was no disparity between tap and bottle drinkers in the other two counties.)

The two lead researchers, Shanna Swan, now a research scientist at the University of Missouri,
Columbia, and Kirsten Waller, now a senior scientist with the Sequoia Foundation in La Jolla, Calif.,
theorized that something in the water supply of that one county was responsible for the higher rate of
miscarriages. Their follow-up study, published in 1998, found that women who drank five or more
glasses of tap water daily in communities where the level of THMs in the water supply averaged above
75 ppb were about 65 percent more likely to experience miscarriages than women who drank less of that
water or who used a different water source. A THM called bromodichloromethane, the researchers
suggested, was a possible culprit.

Similar studies have not found such a link. In research conducted in the early 1990s, Savitz and his
colleagues at UNC found that the source of a woman's drinking water and the levels of THMs in that
water did not increase the miscarriage rate, and, conversely, that the more water women drank, the lower
their risk of premature delivery or delivering a low-birth-weight baby. Savitz noted, however, that
communities with THM levels above 100 ppb may have experienced a slightly higher rate of miscarriages.
This possibility, coupled with the California findings, is prompting his large-scale study.

One factor that will need to be addressed, Savitz said, is how to measure THM exposure.

"Annual utility-wide averages allow for huge variation in levels over the years and across the utility," Swan
said. For example, THM levels may spike above 100 ppb during the summer because of proliferating
aquatic grasses and other organic materials in reservoirs. But there are no scientific data, even in rats,
showing danger from these brief peaks.

Then there's the shower issue. Michele Lynberg, an epidemiologist with the CDC, said that in addition to
swallowing water with THMs, people absorb significant amounts through the skin: during showers, in the
kitchen when boiling water, swimming in a chlorinated pool. So switching from tap water to bottled water
for drinking may not significantly reduce your exposure. One CDC-funded study found that people had
higher levels of THMs in their blood after taking a 10-minute shower than after drinking a liter (about four
glasses) of tap water.

Further Complications

The difficulty of establishing a normal rate of miscarriage also complicates matters. While about 25
percent of all confirmed pregnancies result in miscarriages, many more occur before a woman realizes she
is pregnant. Some experts have estimated that 50 percent of all fertilized eggs are excreted within days of
conception. Factors proven to increase the risk of miscarriages include smoking, alcohol and drug use,
multiple sexual partners, previous contraceptive use and age. The additional effect that THM exposure
might have is therefore difficult to determine, Savitz said.

In the case of Chesapeake, a city of about 200,000 in the Hampton Roads area, the community was
never out of EPA compliance for THMs, according to health officials there: Levels sometimes rose above
100 ppb but not on average for a whole year. THM concentrations now average 40 ppb, said Nancy
Welsh, director of the Chesapeake Public Health Department.

Likewise, the major water facilities serving the Washington area all report being in compliance with the
EPA standards, with their THM averages for 2001 below the new limit of 80 ppb.

Chesapeake's low levels of THMs are largely the result of its new reverse-osmosis water treatment
process, which removes organic material from the water supply. But the process is too costly for many
cash-strapped municipalities and not always practical for the largest systems. Other methods for reducing
THMs include lowering chlorine levels, which raises the risk of water-borne diseases, and disinfecting the
water supply with ozone or a combination of chlorine plus ammonia. Yet even these measures are not

"We are finding more and more that alternative disinfectants may also have negative impacts" on health,
said Philip Singer, director of the UNC's Drinking Water Research Center. Ozone, for example, creates
its own harmful byproducts and it is less effective at killing microbes.

Any movement away from chlorine will need to be undertaken carefully, the EPA warns at the risk, so
to speak, of throwing out the baby with the bath water.

Christopher Wanjek last wrote for the Health section about anti-aging scams.

                    © 2002 The Washington Post Company