Article 20: Does Mercury Matter?

Experts Debate the Big Fish Question
By JAMES GORMAN

What to eat? Sometimes it seems to be an impossible question. There are several different kinds of fats to worry about, complex and simple carbohydrates, vitamins in and out of the bottle, chickens that run free before they die and those that don't, grass or grain-fed beef.

At least all the experts agree that fish is good for you. It's high in protein, low in fat, with those terrific omega-3 fatty acids.

But then, there's mercury. Everywhere on the planet, fish are accumulating mercury in their tissues, often as the result of airborne mercury that finds its way into rivers and seas. And mercury, in all its forms, is highly toxic.

In fish, it occurs in the form of methylmercury, which is known to damage neurons, particularly developing neurons. The damage seen in humans and animals at high doses is severe. Many studies - though not all - have concluded that low levels can have subtle negative effects as well if certain fish are a major part of the diet.

The Food and Drug Administration and the Environmental Protection Agency have advised that groups most sensitive to methylmercury - women of childbearing age and young children - not eat swordfish, king mackerel, tilefish and shark, all of them having relatively high mercury levels. Neither agency defines "young," perhaps because the brain continues to develop until early adulthood, but with each year the potential hazard of mercury is less.

Furthermore, the F.D.A., which oversees fish sold in stores and restaurants, advises these women and children that 12 ounces a week, of various kinds of fish, is safe. The E.P.A., which is concerned with freshwater fish of the sort anglers bring home, advises the same group to eat no more than six ounces week for adults, or two ounces a week for children.

Researchers started to worry seriously about mercury exposure after several incidents of poisoning in Japan and Iraq. In 1972, about 6,500 people in Iraq were poisoned by methylmercury, which was used in an antifungal treatment for grain seeds. More than 400 died. Some of the treated seeds had been mistakenly turned into flour, and then bread.

The need to find out what low doses of mercury were doing, particularly to developing children, fueled several major studies in the Faeroe Islands in the North Atlantic, the Seychelles in the western Indian Ocean, and New Zealand.

Dr. Roberta F. White, chairwoman of the department of environmental health at Boston University and director of the Boston Environmental Hazards Research Center, has been pursuing the Faeroe Islands study for years. She said children exposed to mercury before birth showed clear effects. "The greater the mercury exposure, the poorer they did" on tests that measured nerve functioning, she said.

The effects were subtle, Dr. White said, not something that would be noticed outside a research project, nothing that would be brought to a doctor's attention. She compared the results to some of the damage cause by low-level lead exposure. Similar negative effects were found in New Zealand and in some smaller studies.

Because of these results, she said, "We needed to reconsider how much exposure to methylmercury pregnant women should have."

And exposures were reconsidered.

In June, the World Health Organization changed the level of mercury consumption considered safe from 3.6 to 1.5 micrograms per kilogram of body weight per day. (A microgram is 0.000000035 of an ounce, and a kilogram is 2.2 pounds.)

The Environmental Protection Agency has set its reference dose level, the exposure considered to be safe, at 0.1 microgram per kilogram of body weight per day. A National Academy of Sciences report, basing its findings on the Faeroe Islands study, confirmed that this level was safe.

The Food and Drug Administration uses a different measurement, and it has come under considerable criticism from activist groups like the Mercury Policy Project. Its reference dose is 0.4 micrograms per kilogram of body weight.

However, said Dr. David Acheson, the F.D.A.'s chief medical officer concerned with the mercury issue, the number is misleading. It was based on data from Iraq, he said, and is out of date for groups at particular risk. In practice, he said, the important number is the 12 ounces of various fish each week. That advisory was issued in 2001 and is still in effect. The advisory, he said, "is designed to keep at-risk individuals below" the E.P.A.'s recommended dose. There are also numerous state health advisories. Melanie Miller, director of communications and marketing for the U.S. Tuna Foundation, an industry group, said: "It's very frustrating for us as an industry. Nobody's telling the same story."

Dr. Acheson said the F.D.A. was working out a joint advisory with the E.P.A. that should be issued by January.

The F.D.A. is also looking into another matter of concern, canned tuna.

Michael Bender, director of the Mercury Policy Project of the Tides Center in Montpelier, Vt., has been a longtime critic of the F.D.A. policies on mercury, and most recently his group's efforts have focused on canned tuna. This is by far the most consumed fish, and he argues that consumers deserve to be warned specifically about its mercury levels. His group tested 48 cans of white or albacore tuna and found mercury levels averaging 0.5 parts per million, significantly higher than F.D.A. data. The F.D.A. and the Tuna Foundation agree that albacore has more mercury than light tuna, which comes from smaller fish, but set the figure at 0.3 parts per million compared with 0.13 for light tuna, Dr. Acheson said. Mr. Bender said consumers, particularly women and children, should be warned separately about tuna, since it is so popular. At 0.5 p.p.m. he said, a woman weighing 132 pounds who ate only 6 ounces of tuna in a week would be getting 1.4 micrograms per kilogram of body weight in that week - too much mercury by E.P.A. standards.

Dr. Acheson said that the F.D.A. was starting more intense testing of tuna and that it would consider whether to say anything specific about canned tuna based on those findings.

This week the agency is conducting meetings of so-called stakeholders - including the Tuna Foundation and the Mercury Policy Project - to discuss F.D.A. policy and advisories on mercury in fish.

Ms. Miller of the Tuna Foundation said that her group thought no new warnings were necessary and that the current F.D.A. advisory adequately protected consumers.

Despite the warnings, the results of one study present a challenge to the notion that low levels of mercury have any negative effect. Dr. Gary J. Myers, a professor of neurology and pediatrics at the University of Rochester Medical Center, followed children in the Seychelles whose mothers ate about 12 meals a week of fish with mercury levels comparable to those in fish eaten by consumers in the United States.

Dr. Myers and other researchers did tests that were sensitive enough to pick up the effects on neurological performance of things like poverty and parents' educational level. They found no adverse effect from prenatal mercury exposure. Dr. Myers, who began studying mercury in Iraq, said that in the course of nine years of study in the Seychelles, "We've really not been able to confirm that there are adverse effects" at consumpton levels far above the average American's.

The differences between this study and the Faeroe Islands study have puzzled other scientists. Dr. White, of Boston University, suggested that measures of mercury in umbilical cord blood correlated better with neurological effects than measures in hair.

Dr. Myers said there was no scientific evidence to suggest that the blood measures were better than the hair measures. Dr. Myers also pointed out that the Seychelles population ate fish, whereas the Faeroes population got a substantial amount of its mercury exposure from whale meat. Few people in the United States eat whale meat, he noted.

The National Academy of Sciences panel considered both studies, and others. Dr. Alan Stern, chief of the bureau for risk analysis in the New Jersey Department of Environmental Protection, was on the panel. He said the findings in the Seychelles and the Faeroes were based on sound research. The panel recommended using the Faeroes study to set acceptable levels of mercury intake because other studies had also found neurological problems, and the panel did not want to use one study that found no effect as a benchmark.

This kind of seeming discrepancy occurs all the time in epidemiology, he said. Some research uncovers deleterious effects of chemicals under scrutiny, and other research does not. "From a public health standpoint," Dr. Stern said, if there are well-done studies showing a danger to the public, the best approach is to be cautious and to "make prudent recommendations based on those studies."