Pesticide Residues in Food

Pesticides are substances or mixtures of substances intended for preventing, destroying, repelling, or mitigating plant or animal pests and may include herbicides, insecticides, fungicides, and rodenticides. More than a billion pounds of pesticides are used in the U.S. each year to control weeds, insects, and other organisms that threaten or undermine human activities (U.S. EPA, 2009). Some of these compounds can be harmful to human health if sufficient quantities are ingested, inhaled, or otherwise contacted (see the Urinary Pesticides indicator). Potential health effects and primary exposure routes vary by chemical. The most common routes of exposure for the general population are ingestion of a treated food source and contact with applications in or near residential sites. Pesticides may also be harmful in the environment when non-target organisms are exposed.

This indicator represents data from the U.S. Department of Agriculture’s Pesticide Data Program (PDP), which measures residue levels for hundreds of pesticides and their metabolites in fruits, vegetables, grains, meat, and dairy products from across the country, sampling different combinations of commodities each year. The analysis examines pesticides currently on the market and also includes continued testing for some persistent and bioaccumulative pesticides that have been banned since the 1970s, such as aldrin/dieldrin, heptachlors, DDT, and DDT's metabolites. PDP data collection began in 1991 and includes both domestic- and foreign-produced commodities. Results are published in annual reports, which include statistics on the number of pesticide residues detected, the number of residues exceeding the tolerance established by EPA for a given pesticide-commodity pair (Code of Federal Regulations, Title 40, Part 180), and the number of residues detected for which no tolerance has been established. This indicator depicts data from 1994 to 2014; data from before 1994 are considered less reliable. Between 1994 and 2014, the number of food samples analyzed per year ranged from a low of 5,771 (1996) to a high of 13,693 (2005), and decreased slightly since 2005.

The percent of samples with no detectable pesticide residues generally increased during the period from 1994 to 2002 (Exhibit 1). Samples with no detects accounted for 38.5 percent of samples analyzed in 1994 and rose to 57.9 percent in 2002. Data for 2004–2012 show a lower percentage of samples with no detects than 2003 data, going from 53.9 percent in 2003 to 23.1 percent in 2007, then increasing to 48.9 percent in 2011; since then the percentage of samples with no detects has decreased to 41.5 percent in 2014. The largest increase in detects in the 2003-2014 timeframe was in those samples with detection of four or more residues. The average percentage over the 12-year period is 12.8, but the value tripled from 7.1 percent in 2012 to 22.3 percent in 2014. These trends in numbers of detections have occurred at the same time as a decrease in the analytical limits of detection for various compounds, as instruments are able to pick up ever-smaller concentrations.

Exhibit 2 illustrates the percentage of samples in which at least one pesticide residue was detected at a concentration exceeding the tolerance established by EPA for a given pesticide-commodity pair. The percentage of samples exceeding EPA tolerance values was 0.05 percent in 1994, peaked at 0.53 percent in 2012, and decreased to 0.36 percent in 2014.

• As Exhibit 1 explains, PDP data showing percent of samples with a given number of pesticides detected from 2002 and earlier cannot be compared directly with data gathered after 2002. (Before 2003, each compound detected was counted separately; beginning in 2003, measurement of a parent compound and/or any of its metabolites was counted as a single detect.) Additionally, PDP has refined its analytical methods in order to measure a greater number of pesticide analytes (both parent compounds and metabolites) and lowered its analytical limits of detection. Therefore, some increases in the percentage of detects may reflect improvements in PDP’s analytical method capabilities.
• PDP does not sample all commodities in each individual survey year, which introduces uncertainty in evaluating changes in the percentage of detects and percentage of samples exceeding tolerances. Therefore, differences in the percent of detections for any given pesticide class might not be due to an increase or decrease in the predominance of detectable residues; these differences might simply reflect the changing nature and identity of the commodities selected for inclusion in any given time frame. In addition, PDP may preferentially target and more frequently sample specific commodities that are more likely to have pesticide residue, which may also introduce bias in evaluating trends over time.
• The indicator provides summary information on pesticide residues on food, but does not evaluate exposure from dietary intake or assess risks to human health and the environment.

Data for this indicator were obtained from a series of annual summary reports published by the PDP (USDA AMS, 1996–2016). These reports are all available from https://www.ams.usda.gov/datasets/pdp. The Food and Drug Administration also collects data (not reported here) on pesticide residues in cooked food that may be a source of chemicals in human diets. These data are available at http://www.fda.gov/Food/FoodScienceResearch/TotalDietStudy/default.htm.