News Coverage
Farm-raised salmon called risky choice
A study found more contaminants than in wild salmon. The industry and federal officials objected
Published January 9, 2004
Farm-raised salmon, a popular fish in American supermarkets, is so contaminated with cancer-causing chemicals that people should eat it no more than once a month, a new study says.
The 14 chemicals studied, most of them pesticides, all are banned in the United States. But they can linger for decades in the environment and in the human body, and are present in the smaller fish that are ground up and fed to salmon raised in giant ocean cages.
Wild salmon have much lower levels of contaminants and are safer to eat, the study found.
The controversial study in today's issue of the journal Science, initially rejected for publication until the authors toned it down slightly, throws yet another wrench into the consumer's search for safe, healthy food.
"You've got mad cow disease. You've got Atkins diet and low-carb. And you've got mercury in fish, and now organics in fish," said David O. Carpenter, director of the Institute for Health and the Environment at the University of Albany and one of the authors of the study.
Salmon has long been recommended for its high level of fatty acids, which are good for the heart and brain, and because it has relatively low levels of mercury, another contaminant found in fish.
Carpenter acknowledged that the bad news about organic contaminants in farm-raised salmon was part of a "very confusing message for the consumer."
The study tested contaminants in 700 wild and farmed salmon bought around the world. It found those farmed in Northern Europe contained the most pollutants, followed by North America and then Chile.
Fishing-industry groups denounced the study even before it was released, and the Food and Drug Administration said farm-raised salmon was safe to eat.
The study "will likely over-alarm people in this country," Eric Rimm of the Harvard School of Public Health, a specialist on nutrition and chronic disease, told the Associated Press. "To alarm people away from fish because of some potential, at this point undocumented, risk of long-term cancer - that does worry me."
One fact not in dispute: The amount of at least one chemical measured in salmon was far below the danger level established by the FDA. (Most of the other chemicals do not have established standards for safe consumption, though all are suspected carcinogens.)
With polychlorinated biphenyls (PCBs), the salmon measured 50 parts per billion or less, while the FDA standard is 2,000 parts per billion.
FDA reassurances
But authors said that standard, set in 1984, was outdated. They based their analysis on the far more stringent guidelines used by the Environmental Protection Agency to recommend numbers of meals per month.
EPA guidelines say that if a person eats fish twice a week, it should contain no more than 4 to 6 parts per billion of PCBs. The Science study found that PCB levels in farmed salmon sold in the United States and Canada averaged about five times that amount: more than 30 parts per billion.
FDA officials rejected the notion that their approach was outdated, citing new agency programs to reduce contaminants in fish feed. They also said their standard came from balancing all health impacts, good and bad, that come from eating fish.
"We do not see a public-health concern here," said Terry Troxell, director of the FDA's office of plant and dairy foods and beverages, referring to the new study. "If anything, we'd like to see people eat more" fish.
'Way too late'
Jane Houlihan, vice president of research at Environmental Working Group, said farmed fish would be an important part of feeding the growing world population but said the FDA was not doing its job of ensuring food safety.
"Industry can continue to fall back on that [standard] as a defense for selling fish that's not safe for people to eat," said Houlihan, whose nonprofit group sounded the alarm on PCBs in salmon last year.
The 20-year-old FDA standard also was criticized by Lynn Goldman, a public-health professor at Johns Hopkins University.
"That doesn't pass any sort of scientific scrutiny," said Goldman, the EPA's former assistant administrator for toxic substances.
Carpenter, one of the study's authors, said the health dangers went beyond the cancer risks analyzed in the study, warning that a pregnant woman could harm the brain development of her developing baby by eating too much salmon.
He criticized the advice given to pregnant women to restrict fish intake as "way too late," because the contaminants linger so long in the human body. Some have a half-life of a decade.
"If a little girl that's 10 years old eats salmon [containing these contaminants], when she's 20 she's going to have half that" in her body, he said.
Other sources of beneficial omega-3 fatty acids, besides fish, are flax seed, used in some health cereals, and canola oil, Carpenter said.
In the United States, more than 90 percent of the salmon consumed is "farmed," raised in floating pens, and available year-round. Wild salmon is generally available June through October. Farm-raised salmon sells for about $5 a pound, compared with $15 for wild salmon, said Alex Trent, executive director of the trade group Salmon of the Americas, in Princeton.
The good news is that more than half of the salmon in U.S. stores comes from farms in Chile, which measured lower than most farmed salmon in contaminants. Still, the study recommended one meal a month of Chilean farmed fish.
Mike Gallo, a toxicologist at Rutgers University who has seen the study, said the measurement data were sound, but he likened the advice on meals per month to a "political polemic."
Gallo noted that many of the same contaminants were found in beef and other protein sources.
The study's authors said the levels of contaminants in farmed salmon were two to 10 times higher than in beef.
But Gallo pointed out that people ate far more beef, and estimated that less than 5 percent of human exposure to PCBs came from fish.
"By publishing this polemic, I think it diminishes the impact of Science as a first-line journal," Gallo said. "This type of commentary belongs in a commentary section."
Gallo acknowledged that he was on an experts list prepared by Salmon of the Americas but said he did not know that until this week and was not paid to speak on behalf of the industry.
Trent also said the industry had taken steps to reduce the PCBs and other feed contaminants since the study samples were taken.
The study was funded by the Pew Charitable Trusts.