There were just over 4 million live births in the U.S. in 1992
(4,065,000, to be exact), according to the STATISTICAL ABSTRACT OF THE
UNITED STATES 1997. In addition to these live births, there were
30,000 fetal deaths in 1992, the most recent year for which we have
data. A fetal death is one that occurs after at least 20 weeks of
gestation in the womb but prior to birth. In actuality, there were very
likely more than 30,000 fetal deaths in 1992. The STATISTICAL ABSTRACT
(table 124) says, "There is substantial evidence that not all fetal
deaths for which reporting is required are reported." In any case, life
expectancy at birth in the U.S. in 1992 was 75.8 years, so fetal
deaths that year resulted in the loss of at least 30,000 x 75.8 = 2.27
million person-years of life. In addition, of course, many of these
30,000 fetal deaths precipitated a personal crisis for the parents.
In addition to fetal deaths, there are spontaneous abortions --
pregnancies that terminate spontaneously before the end of the 20th
week of gestation. These are far more common than fetal deaths, though
the exact number is not known. Various studies estimate that
spontaneous abortions occur in somewhere between 6.5% and 21% of all
pregnancies. Thus in 1992, there may have been at least 265,000 to
855,000 spontaneous abortions in the U.S.
Together, spontaneous abortions and fetal deaths are termed
Recent studies indicate that some miscarriages --as well as some
serious birth defects --may be caused by the chlorine added to drinking
water as a disinfectant.
In the U.S., chlorine is added to public drinking water supplies as a
public health measure to kill harmful bacteria in the water. The added
chlorine reacts with naturally-occurring organic matter in the raw
water (chiefly humic and fulvic acids), creating a host of chlorinated
chemicals as by-products. Health agencies, including the federal EPA
[Environmental Protection Agency] simply ignore most of these by-
products and know almost nothing about them. Instead, they focus on
four by-products, allowing these four to act as surrogates for all the
others. The four that EPA pays attention to are chloroform, bromoform,
bromodichloromethane, and chlorodibromomethane. Together, these four
are called "trihalomethanes" or THMs. According to federal drinking
water regulations, if a public water supply serving over 10,000 people
contains more than 100 parts per billion (ppb) of total
trihalomethanes, the water is unacceptable. However, since there are
usually no other available sources of drinking water, EPA is usually
not in a position to do anything except urge the water supplier to try
to clean up its act.
A study by the California Department of Health published in March,
1998, tracked the drinking water consumption and the pregnancy outcomes
of 5144 pregnant women in a prepaid health plan during the period 1989-
1991. This was a prospective study --the drinking water consumption
of the women was ascertained as soon as their pregnancy was registered
in the study's database. Later, the outcome of their pregnancy was
compared with the amount of water they drank and the total amount of
trihalomethanes they received by drinking water (information received
from the water companies). The study found that 16% of women drinking 5
or more glasses of water per day containing more than 75 ppb THMs had
miscarriages, whereas only 9.5% of women drinking less water, or water
lower in THMs, had miscarriages. Thus among women with high exposure to
THMs in drinking water, the likelihood of spontaneous abortion was 1.8
times as great as it was among women with low exposure. Furthermore,
spontaneous abortion occurred, on average, a week earlier among women
with high exposure (10.2 vs. 11.2 weeks of gestation). The strength of
this study was its prospective nature; it did not rely on women to
remember how much water they drank in the past.
To see if their results represented a real effect, the researchers
compared women who filtered their water, or who let the water stand
before drinking it, with women who drank it straight from the tap.
(THMs are volatile and will slowly leave water that is allowed to
stand.) The results were consistent with THMs causing spontaneous
In January of this year, the Agency for Toxic Substances and Disease
Registry published a case-control study showing that serious birth
defects --spina bifida, or neural tube defects --are associated with
total trihalomethanes ingested in drinking water. Neural tube
defects are serious birth defects in which the spinal cord is not
properly enclosed by bone.
This statewide study in New Jersey found a doubled risk of neural tube
defects among those with the highest exposures to THMs in drinking
water. This study pointed out that exposure to THMs can also occur
through the contamination of indoor air. Flushing toilets, showering,
and washing dishes and clothes, can inject THMs into household air,
A previous study of 75 New Jersey towns by Frank Bove had examined
80,938 live births and 594 fetal deaths that occurred during the period
1985-1988. This study examined public water company records and
compared pregnancy outcomes to the amounts of THMs delivered to the
home in drinking water. It did not examine the amount of water
ingested. The study found no relationship to fetal deaths, but the
likelihood of neural tube defects was tripled by exposure to THMs at
levels exceeding 80 parts per billion.
This study provoked a letter to the editor of the AMERICAN JOURNAL OF
EPIDEMIOLOGY, in which the authors suggested a biological mechanism
by which trihalomethanes might cause neural tube defects. Neural tube
defects are known to be associated with vitamin B12 deficiency and the
letter pointed to studies showing that vitamin B12 use by the body can
be disrupted by chloroform, one of the four main trihalomethanes in
chlorinated drinking water.
An even earlier case-control study reported on pregnancy outcomes among
women who delivered babies at Brigham and Women's Hospital in Boston
during the years 1977-1980. Indicators of water quality were taken from
public water supply companies. No data were available on the amount of
water ingested. The water quality indicators were compared among 1039
cases of babies born with birth defects, 77 stillbirths, and 55
neonatal deaths (babies that died within a week of birth) vs. 1177
controls. Stillbirths were 2.6 times as common among women exposed to
chlorinated surface water, compared to controls whose water was
disinfected with chloramine instead of chlorine.
More recently, a study of drinking water and pregnancy outcomes in
central North Carolina reported a 2.8-fold increased likelihood of
miscarriage among women in the highest exposure group for
trihalomethanes in drinking water.
Very recently, a second study from the California Department of Health
has shown that, in one area of California, women who drank cold tap
water had nearly a five-fold increased risk of miscarriage, compared to
women who drank mostly bottled water very low in trihalomethanes.
Bottled water is often disinfected by a process called ozonation
instead of chlorination. Bubbling ozone through water kills bacteria
effectively, avoids the distinctive taste and odor of chlorine in the
treated water, and produces no dangerous trihalomethanes. Many people
buy bottled water simply to avoid the taste of chlorine.
U.S. EPA is currently setting new standards for trihalomethanes in
drinking water. The new regulations would apply to all water companies,
not just those serving 10,000 people or more, and they would limit
total THMs to 80 ppb, down from the present 100 ppb. Still, since
several studies link trihalomethanes at 75 ppb or even less to
increased miscarriages, EPA's new standard seems dubious even before it
has been established.
American water suppliers seem stuck on chlorination as the best way to
disinfect drinking water. However, many European cities, and some
Canadian cities, such as Ottawa, have long ago turned away from
chlorination in favor of ozonation to disinfect their water. In recent
years, a few smaller American cities have begun to use ozonation:
Emporia, Kansas and Littleton, Massachusetts, for example. The Santa
Clara Valley Water District in California has announced that it is
switching to ozonation over the next 5 to 8 years, as has the city of
Las Vegas, Nevada.
Still the vast majority of water supplies in the U.S. remain
chlorinated. And water quality experts remain in the dark about
trihalomethane levels in water delivered to customers. Kellyn S. Betts,
writing in ENVIRONMENTAL SCIENCE & TECHNOLOGY quotes the EPA official
in charge of the new THM regulations saying no one knows how many U.S.
water systems deliver water with THMs exceeding 75 ppb. Betts says
the American Waterworks Association confirmed for her the absence of
data on THM levels in U.S. drinking water systems. The current
reporting system only keeps track of water systems that exceed 100 ppb
as an annual average.
Erik Olson, a water quality expert with the Natural Resources Defense
Council (NRDC), an environmental group in New York City, points out
that THM levels in water supplies typically increase by as much as a
factor of 1.5 to 2 during the summer months. And he says short-term
exposures may be very important in producing some of the pregnancy
outcomes reviewed here --spontaneous abortions, fetal deaths, and
serious birth defects. "We may be totally overlooking the risk of
short- term exposure," Olson said.
--Peter Montague (National Writers Union, UAW Local 1981/AFL-CIO)
 STATISTICAL ABSTRACT OF THE UNITED STATES 1997 [117th edition]
(Washington, D.C.: U.S. Government Printing Office, October, 1997). See
 The STATISTICAL ABSTRACT for 1997, cited above, Table 123, says
there were 7.4 fetal deaths per 1000 live births in 1992, so the total
number of fetal deaths that year was 7.4 * 4,065 = 30,000.
 STATISTICAL ABSTRACT for 1997, cited above, table 117.
 S. Hamamah and others, "The effect of male factors in repeated
spontaneous abortion: lessons from in-vitro fertilization and
intracytoplasmic sperm injection," HUMAN REPRODUCTION UPDATE Vol. 3,
No. 4 (July 1997), pgs. 393-400.
 Kirsten Waller and others, "Trihalomethanes in Drinking Water and
Spontaneous Abortion," EPIDEMIOLOGY Vol. 9, No. 2 (March 1998), pgs.
 Judith B. Klotz and Laurie A. Pyrch, A CASE-CONTROL STUDY OF NEURAL
TUBE DEFECTS AND DRINKING WATER CONTAMINANTS (Atlanta, Ga.: Agency for
Toxic Substances and Disease Registry, January, 1998).
 Frank L. Bove and others, "Public Drinking Water Contamination and
Birth Outcomes," AMERICAN JOURNAL OF EPIDEMIOLOGY Vol. 141, No. 9 (May
1, 1995), pgs. 850-862.
 Andrew T. L. Chen and others, "RE: 'Public Drinking Water
Contamination and Birth Outcomes,'" AMERICAN JOURNAL OF EPIDEMIOLOGY
Vol. 143, No. 11 (June 1, 1996), pgs. 1179-1180.
 Ann Aschengrau and others, "Quality of Community Drinking Water and
the Occurrence of Late Adverse Pregnancy Outcomes," ARCHIVES OF
ENVIRONMENTAL HEALTH Vol. 48, No. 2 (March/April 1993), pgs. 105-113.
 David A. Savitz and others, "Drinking Water and Pregnancy Outcome
in Central North Carolina: Source, Amount, and Trihalomethane Levels,"
ENVIRONMENTAL HEALTH PERSPECTIVES Vol. 103, No. 6 (June 1995), pgs.
 Shanna H. Swan and others, "A Prospective Study of Spontaneous
Abortion: Relation to Amount and Source of Drinking Water Consumed in
Early Pregnancy," EPIDEMIOLOGY Vol. 9, No. 2 (March 1998), pgs. 126-
 Kellyn S. Betts, "Miscarriages associated with drinking water
disinfection byproducts, study says," ENVIRONMENTAL SCIENCE &
TECHNOLOGY [ES&T] April 1, 1998, pgs. 169A-170A.
Descriptor terms: drinking water; trihalomethanes; chloroform; fetal
deaths; miscarriages; statistics; bromoform; bromodichloromethane;
chlorodibromomethane; thms; california; ca; atsdr; nj; new jersey;
neural tube defects; spina bifida; birth defects; epa; ottawa, cn;
emporia, ks; littleton, ma; santa clara, ca; las vegas, nv;