Pregnant women in their first trimester who drink five or more glasses of cold tap water daily may be at higher risk of miscarriage, according to a study of California chlorinated drinking water.

Source: Press release of Reuters press agency, February 11, 1998.



The Californian study

Pregnant women in their first trimester who drink five or more glasses of cold tap water daily may be at higher risk of miscarriage, according to a study of California chlorinated drinking water.

The heightened risk is linked to exposure to a contaminant found in chlorinated water in a majority of municipal water systems nationwide. The chemical -- trihalomethane (THM) -- forms when chlorine reacts with acids from plant material.


Parts of this page

Chlorination byproducts and miscarriages
Comment on the study



Chlorination byproducts and miscarriages

The study, led by California health department investigators Kirsten Waller and Shanna Swann, examined the records of 5,144 pregnant women from the Fontana, Santa Clara and Walnut Creek areas.

The study, which will be published in the Feb. 18 issue of the journal Epidemiology, found that women who drank five or more glasses of tap water per day with at least 75 micrograms per liter of THMs had an increased risk of miscarriage.

Their risk was calculated at 15.7 percent, compared with 9.5 percent among women who received low exposure. Only about 2 percent of the women were exposed to the highest risk levels, meaning they consumed five or more glasses with at least 75 micrograms per liter.


Comments on the study

With a few hours of search, we found evidence that nitrates in drinking water, at only twice the maximum contaminant level (MCL) established by the EPA, can cause 100% miscarriage. Some US states report frequent exceedings of the MCL...

As THM's levels are directly linked to levels of organic contaminants and nitrate levels are connected to fertiliser use and organic wastes, we suppose that THM levels and nitrate levels are positively correlated.

If the California researchers didn't look at nitrate levels and only at THMs, then we are afraid that they looked at the wrong contaminant...

about nitrates and miscarriages

Thrive online says:

Veterinarians have known for years that high levels of nitrate, the primary ingredient in fertilizer as well as human sewage, can trigger miscarriage in hogs, says Greg Steele, an epidemiologist with the Indiana Department of Health. And scientists recognized two decades ago that infants can suffer "blue baby syndrome" when high levels of nitrate impair their blood's ability to carry oxygen. In 1977 concern about that risk led the federal government to set a standard for city water systems of no more than ten milligrams of nitrate per liter of water.

Those regulations, though, don't cover private wells. And when LaGrange County health officials realized that three women who had suffered repeated miscarriages all drank well water, and all lived downstream from a hog farm's leaky septic system, they began to suspect that nitrate might endanger a developing fetus as well as an infant. Sure enough, when Steele and his colleagues checked each woman's well, they found nitrate levels above 19 mg/l--compared to an average of 3.1 mg/l in nearby wells of women who had delivered healthy babies during the same two-year period.

About nitrate in water and infants

The USGS has some comment on nitrate levels:

The EPA has established a maximum contaminant level (MCL) of 10 mg/L (as nitrogen) for nitrate in drinking water. Excessive nitrate can result in restriction of oxygen transport in the bloodstream. Infants under the age of 4 months lack the enzyme necessary to correct this condition. Fatalities from methemoglobinemia ("blue baby syndrome") occur infrequently and are most common in rural areas. Illness and death caused by methemoglobinemia are not always recognized; therefore, its occurrence might be underreported. Although one case in Colorado was attributed to infant formula made from public-supply water that had a nitrate concentration of 13.3 mg/L, most cases involve concentrations that are somewhat higher. In parts of Eastern Europe where ground water is contaminated with 50-100 mg/L nitrate, pregnant women and children under 1 year of age are supplied with bottled water.

About exceeding the 10 mg/l limit

Again the USGS gives places where the limits are exceeded:

A seasonal pattern is apparent in exceedances of the drinking-water standard for nitrate (10 mg/L) in the public water supply of Des Moines, Iowa, from 1974 through 1990. The source of this water is the Raccoon River. Much of the land upstream is planted in corn, soybeans, or other row crops, and fertilizer is heavily used. Exceedances occurred most often during April through July, which is after fertilizers are applied and when storm runoff is frequent (data from Keith Lucey, U.S. Geological Survey, Iowa City, Iowa).

[note: from a graph on that page it can be seen that during the peak months levels exceed the 10 mg/l level average 20% of the days]

About California

The USEPA comments on water quality in California

The State also reports that trace inorganic elements, flow alterations, and nitrates degrade over 1,000 square miles of ground water aquifers.

About chloroform and toxicity

Chloroform, the most abundant THM after chlorination is only toxic at very high levels, not at environmental levels...

Toxicity Data of chloroform
LD oral (rat): 908 mg/kg
LD skin (rabbit): >20 g/kg
LC inhal (rat): 9937 ppm (47,702 mg/m3; 4 h)
PEL (OSHA): 50 ppm (240 mg/m3; ceiling)
TLV-TWA (ACGIH): 10 ppm (48 mg/m3)
Source: HHMI lab safety data sheets

About THMs and pregnancy

The NIEHS has made an investigation in North Carolina:


In spite of the recognition of potentially toxic chemicals in chlorinated drinking water, few studies have evaluated reproductive health consequences of such exposure. Using data from a case-control study of miscarriage, preterm delivery, and low birth weight in central North Carolina, we evaluated risk associated with water source, amount, and trihalomethane (THM) concentration. Water source was not related to any of those pregnancy outcomes, but an increasing amount of ingested water was associated with decreased risks of all three outcomes (odds ratios around 1.5 for 0 glasses per day relative to 1-3 glasses per day, falling to 0.8 for 4+ glasses per day). THM concentration and dose (concentration amount) were not related to pregnancy outcome, with the possible exception of an increased risk of miscarriage in the highest sextile of THM concentration (adjusted odds ratio = 2.8, 95% confidence interval = 1.1-2.7), which was not part of an overall dose-response gradient. These data do not indicate a strong association between chlorination by-products and adverse pregnancy outcome, but given the limited quality of our exposure assessment and the increased miscarriage risk in the highest exposure group, more refined evaluation is warranted.

Key words: chlorination, low birth weight, preterm delivery, spontaneous abortion, trihalomethanes. Environ Health Perspect 103:592-596 (1995)

About chloroform and cancer

The CIIT has done research at what levels chloroform causes cancer and below what level there was no cancer risk left:

Needed Changes in Carcinogen Testing and Risk Assessment: Lessons Learned from Research with Chloroform

May 1994

Chloroform is produced in trace amounts in the process of chlorinating drinking water (in the range of 1 to 50 parts per billion [ppb]). It is also formed as a by-product in some industrial processes, such as the bleaching of paper. Thus, it is important to define the health risks associated with continual low-level exposures to this chemical.

Using the mouse liver cancer incidence data from the gavage study, the linearized multistage risk model used by regulatory agencies to extrapolate from high to low exposures suggests that a 1-in-100,000 lifetime increased risk of cancer would occur with 4 ppb chloroform in the drinking water. This model is very conservative because it assumes that a cancer risk is associated with even the most minute intake of chloroform. In fact, no cancer or increased cell proliferation was observed in the drinking water study, with 1,800,000 ppb of chloroform. Clearly, the conventional model dramatically overestimates risk, and more realistic risk assessment approaches, using more scientific information, are needed for chemicals such as chloroform.



Any natural water contains hundreds of contaminants, including natural organics, salts, harmful bacteria, viruses, etc... Chlorination reduces the acute risk of being killed by infections considarably, but at the same time forms byproducts which "can" be harmful in the long run.

Specifically on the point of spontaneous abortion, we didn't find evidence that THMs, even in large doses, cause this phenomen in animal trials. On the other hand, there is evidence that other contaminants, specifically nitrates, are involved in that problem in animals and humans in doses less than twice the today's imposed limits.

Children below 4 months old do suffer from suffocation above certain nitrate levels in food and drinking water. Because nitrates just pass the placenta without any problem, there is reason to believe that this too can result in killing the child in the womb.

Objections against this kind of research, is that widespread publication of such partial or improper results can introduce a lot of unnecessary fears and even a lot of damage. Should we remind you of the Peru cholera disaster, where officials stopped chlorination of drinking water, amongst others based on USEPA research "linking" chlorination byproducts like chloroform to certain cancers?


You are at level one of the Chlorophiles comment pages

Created: February 16, 1998.
Last update: September 6, 1998.

Welcome page

To the Home Page of the Chlorophiles

Left: PVC and cancer

Right: The Houston Chronicle Articles on VCM

For any comment on this or other pages, especially on chlorination of water:

Mail to: