Filtering the Data on Water Safety
From the publishers of the New England Journal of Medicine
On the long list of things scientists know or suspect may cause cancerfrom cigarettes to sunburns to charbroiled steaksdrinking water seems a most unlikely addition. After all, most tap water looks and tastes clean and pure. Recent reports of a possible cancer risk from chlorination by-products, which are found in more than half of the nation's drinking water, may have raised concerns. But while the studies add more data, they provide no definitive proof.Nearly all water in the US is disinfected with chlorine, which kills nasty microbes that cause typhoid, cholera, and other waterborne infections. But there's an unfortunate side effect: Chlorine combines with decaying plant matter to form chloroform and related compounds called trihalomethanes. These by-products are present only in very tiny amounts, but they tend to be more prevalent in communities that use surface water, which comes from reservoirs or other open sources. Groundwater, in contrast, is pumped from underground water tables and contains less of the precursors that combine with chlorine. Chlorination by-products have been scrutinized for their cancer-causing potential since the 1970s. A number of studies have hinted at a higher risk of bladder and colorectal cancer from the chemicals. This summer, a Finnish study in the Journal of the National Cancer Institute reported that rats given water containing high levels of a chlorination by-product known as MX developed tumors of the thyroid, adrenal glands, and other organs. Given the potency of MX, researchers estimate the chemical may account for as much as 50 percent of the cancer-causing potential of chlorinated water. But they don't know exactly how MX causes tumors, and extrapolating the results to humans is difficult, particularly since the types of tumors seen in the rats are different from those most strongly linked to chlorinated drinking water in humans.
The US Environmental Protection Agency (EPA), which oversees the safety of the nation's water supply, has set the maximum contaminant level for trihalomethanes in drinking water at 100 parts per billion. The agency doesn't routinely monitor for MX, but a handful of studies in the US and Finland suggest that levels in our drinking water are far lower, ranging from 3 to 67 parts per trillion. That means even the highest levels are more than 1,000 times lower than the highest acceptable levels of trihalomethanes. In a worst-case scenario, a lifetime of drinking water with MX levels at that upper range might account for two excess cancers per million people, says Ronald Melnick, PhD, a toxicologist from the National Institute of Environmental Health Science who co-authored an editorial accompanying the MX study. What about other, more well-documented chlorination by-products? A report in the July American Journal of Public Health looked at the link between chloroform in drinking water and cancer in more than 28,000 postmenopausal women in Iowa. Researchers estimated chloroform exposure from mailed surveys that included questions about what type of water the women drank, along with statewide water quality data. Most of their water came from public water systemsgroundwater, surface water, or a combination of the two. After adjusting for other factors known to influence cancer risk, such as age, smoking status, and fruit and vegetable intake, researchers found that the risk of colon cancer rose in tandem with increasing chloroform levels. Compared to the women whose water contained low levels of the contaminant, those with highest levels had a 68 percent higher risk of colon cancer. These women also had a higher rate of all cancers combined. Still, the results don't prove that chloroform is the culprit. As the authors point out, other chlorination by-products or contaminants such as pesticide residues may be stronger carcinogens. Chloroform may simply be a red flag for something else that's the true cause of the increased cancer risk. For example, the women who drank surface water, which tends to have higher chloroform levels, were also more likely to live in urban environments. This might explain their higher cancer rates, points out Tim Ford, PhD, of the Harvard School of Public Health.
Even if drinking water contaminants are to blame, the actual increase in risk is quite small. For example, the odds of a 60-year-old woman developing colon cancer are about 1 in 1,000. Increasing that risk by 68 percent translates to 1.68 in 1,000not even a doubling of the odds. Family history is probably the single most consistent risk factor for colon cancer, increasing the chances of the disease by about threefold, on average. Other suspected risk factorsa lack of exercise and a high-fat, low-fiber dietaren't considered that definitive. Polluted water is not something experts generally worry about, says Arthur Schatzkin, MD, a colon cancer expert at the National Cancer Institute. But if you're concerned, you can have your tap water tested for chloroform. If the levels are high, an activated carbon filter, which helps reduce chlorine and chlorination by-products, might be a worthwhile investment, if only for the peace of mind it provides. Of the five scientists interviewed for this story, only one said he used a filter, but largely because it improves the water's taste. If you use one, replace the filter on schedule so it works properly. Since the EPA doesn't regulate bottled water, that is not necessarily a better choice. Some may be drawn straight from a municipal water supply or sources such as springs that could have other contamination problems. Another option is to let the tap run for a minute or two first thing in the morning or anytime it hasn't been used for several hours. Doing so will help flush out any chlorination by-products that may have formed while water sits in the pipes, says Dr. Ford. Experts agree it's a good idea to know the source of your drinking water and to be vigilant about protecting and conserving local water supplies. But the cancer risk from tap water is probably miniscule. As HealthNews associate editor Harry Greene, MD, points out, the balance of studies shows no direct link between current levels of water chlorination and human cancer, and in the few studies that suggest a link, the risk is extremely small and unconfirmed. J.Corliss For More Information