STATS at George Mason University

Non-cancer health risks

Despite all these objections, one of the authors of the study, David Carpenter, Director of the Institute for Health and the Environment at the University of Albany, told the BBC (Newsnight, 08/ 01/04) that: “our cancer risk assessment is in fact an underestimate of the true risk, because we have only dealt with cancer risks from three of the 14 contaminants, and we have not yet been able to do a risk assessment with the non-cancer health effects of these contaminant, which are very significant.”

The non-cancer health risks consist of developmental problems in fetuses and children, and endocrine and thyroid problems. These have long been a concern to scientists and regulatory bodies, given that wild fish caught around PCB hotspots in lakes and rivers may contain considerably higher levels of PCBs (over 3000 parts per million) than farmed salmon.

Dr. Carpenter went so far as to tell the Vancouver Sun that women of childbearing years should, as the paper put it, “be particularly wary of eating farmed salmon from any source because of the threat that PCB accumulations pose to reproductive systems and intellectual development.” Similar statements appeared in many other news accounts; but nowhere was it mentioned that the EPA had calculated a much lower risk for these problems — four times lower — than for cancer.

The Vancouver Sun did balance Dr. Carpenter’s warnings with comments from John Salminen, chief of chemical health hazard assessment at the health products and food branch of Health Canada, the Canadian equivalent of the FDA. “Salminen... described Carpenter’s suggestion of health risks for women and children as ‘alarmist.’ He said a researcher could portray any protein source, including fish, cows, pigs and chickens, as unsafe simply by applying a sufficiently restrictive set of contaminant standards... ‘If you take a conservative enough approach and take an excessively high uncertainty factor you can conclude that you should not eat anything that has any fat content.’”

Problems with the sample of salmon
In the days that followed publication, questions were raised about exact provenance of the sample of fish from Scottish salmon farms, which recorded the highest levels of PCBs in the study. An editorial in the Scotsman (16/01) pointed out that the sample had been purchased in March 2002, when supermarket fish sold in the United Kingdom was not labeled as to its origin. “... the scientists do not actually know if they tested Scottish farmed salmon,” claimed the paper, adding, “Indeed the American scientists concerned were blithely unaware that you could buy wild salmon in a Scottish fishmonger’s or supermarket.”

Another consequence of analysis conducted on fish bought in 2002 is that the PCB levels may well be out of date. Industry efforts to reduce PCBs in fish resulted in a decline of 28 percent between 1998 and 2001 (the most recent FDA figures). America’s salmon industry association “Salmon of the Americas” claims that industry-wide sampling has shown a further reduction of 30 percent since 2001. The replacement of fish oil feed with vegetable oil feed will bring about further substantial reductions.

about staff publications email contact search contribute
Salmon home
email this page to a friend

STATS home