• Increase font size
  • Default font size
  • Decrease font size
Home Is Our Fish Fit to Eat?

Is Our Fish Fit to Eat?

E-mail Print PDF


A six-month investigation of fresh fish and shellfish raises serious questions about their quality.

Americans are eating more fish. Over the last decade consumption per person in the U.S. has risen nearly 25 percent - from 12 ½ to 15 ½ pounds a year. However, a six-month investigation by CU raises questions about the quality, wholesomeness, and, in some cases, the safety of the fish consumers are eating. Almost 40 percent of the fish we sampled was of fair or poor quality, and 30 percent was downright poor.

Nearly half the fish we tested was contaminated by bacteria from human or animal feces, most likely the result of poor sanitation practices at one or more points in the fish-handling process. Some species were contaminated with PCBs and mercury which can pose health hazards to certain people (see What Else Is in Fish?, page 112).

We also found that many retailers were engaging in deceptive selling practices. Some merchants refer to the fish in their counters as "fresh" when it has actually been frozen and then thawed for display. Others mislabel the species of their fish, either deliberately or out of ignorance. As a result, consumers end up eating some fish other than the one they thought they had purchased, usually paying more than they should have. One-third of our fish samples were mislabeled.

To check for fish quality and chemical contamination, we bought seven popular and readily available species – salmon, flounder, sole, catfish. swordfish, lake whitefish, and clams-from supermarkets and fish  stores in New York City. Chicago. and their suburbs. Except for clams, which were in their shells, we always purchased steaks and fillets. the cuts most people buy.

Our reporter, a CU shopper, and an expert on fish quality visited stores, observed display counters. and purchased samples. We bought whatever fish on our list the store had available: we did not search out fish of obvious poor quality.

We shipped our 113 samples in refrigerated containers by overnight delivery to a contract laboratory. The samples were all in their original store wrappings when they arrived at the lab.

To check the accuracy of package labeling, we bought a different set of samples in metropolitan New York, metropolitan Chicago, and in the San Jose-Santa Cruz area of California. A report on those results begins on page 110.

How good fish goes bad

Fish is one of the most perishable of all foods. Bacteria are the main culprits in the spoilage process, although natural enzymes that break down the flesh of dead fish and oxygen in the air do their part, too. Then ocean fish begins to spoil, it produces a compound called trimethylamine, which causes the ”fishy” odor that people associate with bad fish. Fresh ash has virtuaIIy no odor.

Tine and temperature are the enemies of freshness and flavor. Bacteria multiply rapidly when fish are out of "water too long or are stored at temperatures that are too high. Since fish are cold·blooded and live in cold environments, many kinds of bacteria that live on them also thrive at colder temperatures. As a result, those bacteria are quite comfortable at temperatures found in and, guidance branch at the FDA's Office of Seafood. "We go after blatant violators and hope to set an example."

Fishermen at Sea with nets of fish
Net worth
The U.S. fishing industry hauls in about 9.4 billion pounds of fish each year, worth roughly $3.5- billion. More than half the catch goes for human consumption, the rest for pet food and industrial products.










State regulators are also supposed to keep an eye on food labeling in supermarkets. We didn't find them overly concerned about counterfeit fish. Many state agencies have barely enough money to conduct sanitation inspections, let alone worry about fish labels. In New York, food inspectors don't even look for misbranded fish at every routine inspection.

In Illinois, inspectors from local health departments inspect retail grocery stores. They don't look for mislabeling, either. Says Kerry O'Shaughnessy, a health inspector for the village of Glenview, "I know red snapper has kind of reddish skin, but I couldn't tell you if the store substituted something else."



The waters where fish live are often dumping grounds for potentially harmful chemicals that have been used on land.

Once in water, these substances make their way into the sediments at the bottom and into aquatic plants and animals at the base of the food chain. Little fish eat plants and little animals, bigger fish eat the little fish, and so on up the food chain. Fish also absorb these substances directly from water that passes over their gills. Older fish, predatory fish, and fatty species of fish accumulate more such substances in their tissues than younger, smaller, or leaner ones do.

Although the Environmental Protection Agency banned polychlorinated biphenyls (PCBs) and most chlorinated hydrocarbon pesticides in the 1970s, some residues that were released into the environment before these compounds were banned still remain. They do not decompose easily, so they linger in the water sediments and in the tissues of fish and humans for years.


Fish/Key findings

SALMONsalmon: Forty-three percent of our samples contained PCBs, a potential carcinogen and reproductive hazard.

swordfishSWORDFISH: Ninety percent contained mercury, which may harm the nervous system. Twenty-five percent contained PCBs.

catfishCATFISH: An occasional sample contained residues of the pesticides DDT, DOE, and DOD, which can affect reproduction in mammals.

clamsCLAMS: Some samples were high in lead, which can impair behavioral development in young children.


Lake WhitefishLAKE WHITEFISH: Fifty percent of our samples contained PCBs. Some contained traces of pesticides.


Flounder - Sole FLOUNDER, SOLE: Our flounder and sole were virtually free of pollutants. Fifty-five percent had no detectable residues; the rest, barely detectable. In addition to our lab tests for bacteria, we analyzed all of our fish samples for PCBs, mercury, and for a long list of pesticides. 'We also looked for lead, cadmium, and arsenic in clams.

Forty-three percent of our salmon, 50 percent of our whitefish, and 25 percent of our swordfish contained detectable levels of PCBs. Ninety percent of the swordfish also contained detectable amounts of mercury. In some of these fish, the levels we found were significant, posing a possible health hazard to developing fetuses.

Here's a rundown of what 'our laboratory tests found:


PCBs are synthetic liquids that were once used in electrical equipment, hydraulic fluid, and carbonless carbon paper.

PCBs promote cancer in laboratory animals, but researchers now believe that a greater hazard may be their effects on human reproduction. One study of women who ate fish contaminated with PCBs from Lake Michigan found they gave birth to smaller babies with significant developmental problems. These women reported eating fish for several years before they conceived.

By far the biggest source of PCBs in the human diet is fish, particularly fatty species, such as salmon. Like humans, all fish have PCBs in their tissues, but some have more than others. The concentration of contaminants in fish is related to the waters they live in.

The FDA established an official tolerance of 2 parts per million for PCBs in 1979.

The tolerance is the agency's basis for taking legal action and destroying contaminated samples of fish.

Although most of our samples were within the tolerance levels set by the FDA and would have passed the agency's inspection, that's scant comfort. As PCBs linger in the environment, their composition changes, and they gradually become more toxic. These "weathered" forms of PCBs are more toxic than those forms tested in studies on which the tolerance was based. And these more toxic forms are likely to be found in fish.

Furthermore, the FDA set its tolerance before the hazards to human reproduction were documented, at a time when people were eating less fish, and before the toxicity of PCB breakdown in the environment was known.

The last factor is especially significant, since PCBs accumulate in body tissue. The PCBs that you eat today will be with you decades into the future.

Given these facts, we think even 1 part per million of PCBs in fish is too high. Our laboratory detected levels ranging from 0.2 to 2.1 parts per million in our whitefish, swordfish, and salmon. Three out of 10 samples of whitefish contained PCBs exceeding 1 part per million; three out of 20 samples of swordfish did.

Seven of 10 salmon samples we purchased in New York contained PCBs ranging from 0.7 to 1.3 parts per million. Thirty percent of the samples from Chicago had detectable levels, ranging from 0.2 to 0.8 parts per million.

Some of our Chicago salmon samples were probably species from the West Coast, at least that's what the store clerks told us. Those salmon may have come from less-contaminated waters than fish from the Atlantic or the Great Lakes, the possible sources of fish we purchased in New York. However, because the package labels, store clerks, and signs were not always believable, we couldn't tell for sure where our fish was from.

Nor could we tell whether it was “farm-raised," as some salmon is. Just because salmon is farm-raised doesn't mean it's contaminant-free. Farm-raised fish spend part of their time in pens in the ocean. Their diet also consists of manufactured feed, which is based largely on fish that may have contained PCBs.

Advised, but not advertised States collect samples of fish from local waterways to test for contaminants. If a body of water is particularly polluted, states issue advisories, generally aimed only at sports fishermen. We found advisories inconsistent from state to state, often offering different advice about the same fish from the same waters.


Mercury is a poisonous metallic element that is released by burning fuels as well as by industrial and household wastes. Eventually it settles in waterways and oceans where it joins naturally occurring mercury. There, bacteria convert it to the toxic compound methylmercury.

Methylmercury is a poison that affects the development of the nervous system. Unlike PCBs, which, linger in the fatty tissues of humans for many years, mercury eventually leaves the body, usually within two years, provided you stop ingesting it .

Mercury accumulates in large fish that live a long time, such as tuna, shark, and swordfish. Almost all of the mercury in fish is the compound methylmercury. The FDA has set an informal action level of 1 part per million for methylmercury in fish.

Ninety percent of our swordfish samples had detectable levels of total mercury, ranging from Q,46 to 2,4 parts per million. The average of those samples was 1.14 parts per million. Forty percent contained mercury that exceeded the action level.

The high levels of mercury we found in swordfish are not surprising, since mercury is present in oceans throughout the world. There's no way to prevent swordfish from accumulating it in their tissues.

The FDA told us, however, it is "actively reevaluating the action level for methylmercury," a step in the right direction. The Canadian Health Protection Board has a lower standard for mercury in fish-O.5 parts per million. Eight-five percent of our swordfish samples would have exceeded the Canadian limits.


Residues of pesticides used on farms and forests, and for mosquito control ultimately find their way into rivers and oceans. There they remain for many years. For example, although the EPA banned the use of DDT in 1972, several of our fish samples had measurable levels of DDT and the products it creates when it breaks down-DDD and DDE-in their tissues.

DDT and its breakdown products, as well as the pesticides dieldrin and endrin, can all affect the human nervous system. DDT, DDE, and dieldrin have also caused liver tumors in rodents, and all of these pesticides affect reproduction in mammals.

The FDA has set an informal action level of 5 parts per million for DDT in fish and 0.3 parts per million for dieldrin and endrin.

Two of our samples, one of catfish and one of sole, contained more than 6 parts per million of DDE. Several others, notably catfish, had lower levels of DDT and its breakdown products. The presence of DDT in catfish is not surprising, since they are often raised in ponds on land once used for agriculture.

Our lab reported levels of the pesticide dieldrin over the action level for two samples of lake whitefish and levels of endrin from 0.1 to 0.2 parts per million for four samples of swordfish, salmon, and flounder.

The species most free of pesticides were flounder, sole, and clams. We detected no pesticide residues in half of our New York flounder samples; the other half contained only trace amounts. Six of the 10 sole samples purchased in Chicago had no detectable levels of pesticides; the others contained levels of DDE ranging from a trace to more than 6 parts per million.

DDT and its breakdown products are widely dispersed in the environment and will show up in the human diet for decades to come. While its use has been banned in the U.S., DDT is still used in other countries. However, since only an occasional sample of our fish had significant pesticide residues, the overall hazard from this group of contaminants appears relatively low.

What's in clams?

We found no measurable levels of pesticides in clams. But a number of our samples had relatively high levels of arsenic and lead. The hazards posed by arsenic in shellfish are not clear. Lead, even at very low levels impairs behavioral development in young children. The FDA has set no action level for lead in seafood.

About half of our samples contained lead ranging from 0.31 to 7.8 parts per million. Compared with other foods, such as fruits and vegetables, which typically contain 0.01 parts per million, the lead levels in clams are high. •


Last Updated on Thursday, 17 December 2009 05:13