the film forum library tutorial contact

Questions Raised Over Salmon/PCB Study by Bill Rudolph NW Fishletter, August 15, 2003

A July 30 report that said farmed salmon contain five to 10 times as much PCBs as wild salmon has created a worldwide stir over health concerns. But now that more than 75 versions of the fish story have appeared, ranging from the New York Times to the BBC and even showing up on web-based news sites in Saudi Arabia, it seems likely the major media have been duped once again.

The report, conducted by a Washington DC-based consumer organization called the Environmental Working Group, sampled farmed fish from 10 US stores. It reported PCB levels ranged from 5 parts per billion in Chilean farmed fish to nearly 70 ppb in fish from Scotland, with most US and Canadian samples at around 25 ppb. But the EWG report failed to mention that two studies from the late 1990s found that wild salmon contained higher levels of PCBs than the farmed salmon EWG analyzed in last month's scary report, which was not peer-reviewed.

Polychlorinated biphenyls, a class of chemical compounds once used extensively as industrial coolants and in electrical transformers, are suspected of causing cancer in humans at some level. They were banned in 1976, but they remain throughout the environment.

Wild fish get PCBs through the food chain, since the contaminant is highly dispersed in the oceans, especially the northern hemisphere. Some species of salmon, like chinook, that feed at higher trophic levels than chum or sockeye because they eat other fish like herring instead of plankton and tiny shrimp, generally have higher levels of PCBs, since the contaminants typically concentrate by a factor of 10 from one trophic level to the next.

But farmed salmon, which spend their lives in pens, get PCBs from fish meal made from oily fish like herring. The contaminants tend to concentrate in fatty tissue.

However, a 1998 report issued by the Washington Department of Fish and Wildlife found levels of PCBs in Puget Sound chinook comparable to the highest found in farmed salmon in the EWG's report, levels around 74 parts per billion from 34 chinook caught in marine areas, and 49.1 parts per billion for 144 other chinook caught in rivers. Thirty-two coho salmon tested in marine areas averaged 35 ppb.

In another study, Swedish scientists reported in 1998 in the peer-reviewed journal Arctic that wild Copper River sockeye in Alaska contained PCBs at levels of nearly 700 parts per billion (measured by lipid concentration) while they were still in the ocean. The article said the salmon transported the contaminants to lakes where they spawned; the PCBs were then picked up in the diets of fresh water fish like grayling.

NOAA researcher John Stein said dividing the 700 ppb by a factor of 10 would approximate the "wet weight" measurement used in the other studies, which puts the Copper River fish close to Puget Sound chinook in terms of PCB levels.

The EWG researchers didn’t measure PCB levels of wild salmon themselves, but used data reported in an earlier pilot study sponsored by the David Suzuki Foundation, an environmental group based in Vancouver, BC. The pilot study sampled four wild fish and four farmed fish, and its results were published in the peer-reviewed journal Chemosphere in 2002 (Easton et al). The 12-pound wild chinook in that study had the highest level of PCBs of any of the wild fish tested, about 8.5 ppb. Samples from a chum and two sockeye, which feed at lower trophic levels, tested slightly lower. The Easton study found the farmed salmon had PCB levels that ranged from 20-72 ppb, which are in the same range as PCB levels found in the Puget Sound and Copper River salmon.

The environmental groups sponsoring these studies say that consumption of farmed salmon should be limited because of concerns over the PCBs. Their recommendations have been challenged by Prof. Charles Santerre, an associate professor in the Department of Foods and Nutrition at Purdue University and a spokesman for the Institute of Food Technologists.

In an open letter to Chemosphere after the Easton study was published, Santerre disputed the results on several fronts. He said the sample size was too small, and that the article was confusing and could keep people from eating salmon.

Santerre also pointed out that his own research has shown that farm-raised fish "generally are much lower in contaminants because they are fed a commercial feed rather than having to obtain their food in other manners."

Commenting on the EWG report, he pointed out that salmon species are low in harmful contaminants, "like mercury and PCBs, which are common in recreationally caught fish and some commercial fish."

Santerre said the EWG report "incorrectly uses a more conservative or lower allowable limit for PCBs in fish." He said the main risk from PCBs in the diets of pregnant or nursing women and young children is not from an increased risk of cancer, "but rather, that at higher levels, they may result in developmental delays in the young. While a person consuming farmed salmon weekly over a 70-year lifespan may slightly increase their risk of cancer, the heart-healthy benefits to maintaining a diet rich in long-chain omega-3 fatty acids far outweigh the risks," Santerre said.

"The EWG report also suggests that the consumption of farmed salmon is comparable to the consumption of beef, pork and poultry in the US," he said. "Per capita consumption of salmon is two pounds, while consumption of beef, pork and poultry is 191 pounds combined. Therefore, a person will receive more of their PCBs from foods other than farmed salmon."