Government agencies in charge of monitoring food for pesticide residues must step up their testing programs, said the Government Accountability Office in a new report. While data collected by these agencies has shown low levels of pesticide residue violations in the past few years, shortcomings in sampling methods mean some residue violations may be going undetected, according to the report, published Friday. GAO recommended that the agencies report shortcomings in their methodology to make the significance of their findings clear. Three government agencies — the U.S. Food and Drug Administration, the U.S. Department of Agriculture’s (USDA) Food Safety and Inspection Service (FSIS) and USDA’s Agricultural Marketing Service — are responsible for testing the U.S. food supply for pesticide residues. FDA’s and FSIS’s findings are used to enforce the residue tolerance levels set by the Environmental Protection Agency, while AMS’ data is used to give EPA an idea of dietary exposures to pesticides from year to year. In its report, GAO found that both FDA and FSIS should disclose the limitations of their testing programs in their pesticide residue reports, in accordance with best practices set forth by the Office of Management and Budget. FDA should also examine the possibility of making its sampling program statistically significant, recommends the report. FDA’s sampling program is not statistically valid, the study found, because it targets foods that have already been shown to have higher residue levels in the past rather than selecting samples completely at random. This means that findings of pesticide residues, or a lack of residues, cannot be extrapolated to reflect a greater trend. Without a random sampling method, FDA, which is responsible for testing all produce, dairy products, seafoods and spices, cannot determine the overall incidence of pesticide residues in bigger populations of foods, such as domestic or imported foods, or a certain commodity. The report also found that FDA’s new tool for identifying foods that might be more likely to carry high levels of pesticides is flawed because it relies on multiple sources of information. In addition, FDA does not test for many pesticides for which EPA has set tolerance levels, found GAO, a fact it said the agency should disclose in its reports. In its review of FSIS’s testing, GAO said the agency, responsible for animal products, does have a statistically sound sampling method. GAO noted that FSIS dramatically reduced the frequency of its residue testing between 2000 and 2009, over which time the number of samples collected dropped from 8,000 to under 1,900 per year. Since 2010, however, that number has started to climb up again as FSIS works with EPA to expand its testing capabilities. GAO also noted that, like FDA, FSIS does not test for all pesticides with established tolerances.  After updating its equipment to be able to test for more pesticides, FSIS is now testing for 85 of the 207 substances on EPA’s priority list. Both FDA and FSIS should disclose which pesticides for which they are not testing, despite the fact those chemicals have established tolerance levels, said GAO. The report generally praised AMS’s residue testing program. While the agency has some minor problems with its methodology, according to GAO’s findings, it has a robust sampling program that provides “valuable information on the incidence and level of pesticide residues in food.” GAO acknowledged that the bulk of the pesticide monitoring task falls to FDA. The agency is responsible for monitoring residues in approximately $400 billion worth of domestic foods and $50 billion in imported foods. In response to GAO’s study, FDA said it is expanding its pesticide testing program, but that its top priority is preventing human illness from microbiological contamination in these same foods, so that is where it allocates the bulk of its resources. FDA also noted that implementing a statistically valid sampling program would require significantly more resources, and that it uses data from AMS too identify problem foods to assign certain commodities to a higher risk category. History of Pesticide Residue Testing In 1996, the Food Quality Protection Act mandated that the Environmental Protection Agency reassess all existing tolerance levels for pesticides used on food that had been approved for use before November of 1984. By 2007, EPA had completed 9,721 reassessments, which are now used to determine whether detected pesticide residues are at safe levels. Since the reassessments, the use of certain pesticides, such as organophosphate insecticides, which were shown to be toxic at high levels, have decreased significantly. U.S. farming operations used an estimated 684 million pounds of conventional pesticides in 2007, the most recent year for which data is available. GAO noted that it’s difficult to associate the amount of pesticides used with the risk to human health, since they constitute such a wide variety of chemicals in varying amounts. “We were unable to find publicly available estimates of the overall toxicity or risk associated with the use of agricultural pesticides in the United States,” note the authors.