Compliance Program report of findings: FY 79 pesticides and metals in Fish Program (7305.007)

Because fish bioaccumulate* certain chemicals, levels of chemical contaminants in their edible portion must be closely monitored. In recent years, FDA has conducted several surveys designed to determine the occurrence and levels of selected chemicals or groups of chemicals in fish. Previous fish surveillance programs included the Mercury in Wholesale Fish Survey (FY 71), the FY 73 and 74 Comprehensive Fish Surveys, the Canned Tuna Program (FY 75), the Kepone and Mirex Contamination Program (FY 77), and the FY 77 Mercury in Swordfish Program. In addition, recent Compliance Programs for Pesticides and Metals in Foods and Pesticides, Metals, and Industrial Chemicals in Animal Feed have specified coverage of fish and fish products. Because of previous findings and the sustained high level of fish imported into the United States, a separate compliance program dealing solely with chemical contaminants in fish was initiated by the FDA Bureau of Foods in FY 78. The program includes all domestic and imported fish coverage except that directed by the Bureau of Veterinary Medicine for animal feed components derived from fishery products. The earlier surveys indicated that "bottom feeder" species such as catfish generally had the highest levels of pesticides and polychlorinated biphenyls (PCBs). For this reason, coverage at these species has been emphasized. Similarly, tuna has received special attention because it is the most prevalent fish in the U.S. diet and because of potential problems with mercury. Halibut, swordfish, and snapper also were emphasized in the sampling because of potential problems with mercury levels determined in previous years. The findings in this program were used in detecting emerging problems in fish and directing FDA efforts to deal with them. Care must be exercised in drawing conclusions about trends from the data because this Compliance Program was not statistically designed. Sampling objectives and sources may vary from year to year; thus the results are not directly comparable.