$ < 73 \ Ml (T PRO^ b 2 ui (J T J OFFICE OF INSPECTOR GENERAL Catalyst for Improving the Environment Supplemental Report Details on Dietary Risk Data in Support of Report No. 2006-P-00028, "Measuring the Impact of the Food Quality Protection Act: Challenges and Opportunities" August 1, 2006 ------- Table of Contents Methodology on Approach to Track Risk Mitigation Using Dietary Pesticide Residual Data from U.S. Department of Agriculture's Pesticide Data Program S-1 Aggregate Dietary Risk Index Values for Selected Foods S-7 Impact of EPA Actions on Risk Driver Pesticide-Food Combinations from Domestic Commodities S-9 Impact of EPA Actions on Risk Driver Pesticide-Food Combinations from Imported Samples S-11 Abbreviations aPAD Acute Population Adjusted Doses aRfD Acute Reference Doses cPAD Chronic Population Adjusted Doses cRfC Chronic Reference Concentration cRfD Chronic Reference Doses CRS Chronic Risk Share DRI Dietary Risk Index EPA U.S. Environmental Protection Agency FQPA Food Quality Protection Act PAD Population Adjusted Dose PDP Pesticide Data Program PRLgg Projected 99th Residue Level cRfCSf Single-food Chronic Reference Concentration USDA U.S. Department of Agriculture ------- Methodology on Approach to Track Risk Mitigation Using Dietary Pesticide Residuai Data from U.S. Department of Agriculture's Pesticide Data Program We conducted an analysis of the dietary pesticide residue data from the U.S. Department of Agriculture's (USDA's) Pesticide Data Program (PDP) to evaluate the impact of the Food Quality Protection Act (FQPA) on dietary pesticide exposure risk for children.1 A methodology was developed to track changes in dietary pesticide risk levels between 1994 to 2003 in USDA's pesticide residue data from the PDP. The U.S. Environmental Protection Agency (EPA) regulates dietary risks under the FQPA at the 99.9th percentile level of exposure, based on a probabilistic distribution of dietary exposures. Monte Carlo simulation methods are used to generate hundreds of thousands to millions of "eating day episodes" for a person of known weight. Each "eating day" estimate of pesticide exposure is based on the actual foods reported as eaten in USDA's food consumption survey. Each food reported as eaten is linked to a distinct record in the pesticide residue data file for the same food. The computer randomly selects a residue value from the file, such that the most common levels (usually zero residue) are chosen more frequently, and higher residue levels are picked only as frequently as they appeared in PDP sampling. An estimate is made of a person's daily exposure to a given pesticide by multiplying the amount of each food the person consumed (in kilograms) by the concentration of the residue in each food (in parts per million, or milligrams per kilogram). These estimates for a given pesticide are then aggregated across all the foods the person consumed in a given day, and the results are arrayed from the highest to the lowest, based on the milligrams of pesticide consumed per kilogram of body weight. Under science policies developed to guide implementation of the FQPA, EPA strives to assure that pesticide tolerances are set at levels safe for all population subgroups. Typically, the age group that is exposed to the greatest amount of pesticides per kilogram of body weight is l-to-2-year-old children. Hence, we focused on children's exposure and risk levels in this report. A tolerance level is regarded as acceptable if the child at the 99.9th percentile level2 of the exposure distribution curve is exposed to less of a pesticide than allowed, given the weight of the child and the pesticide's acute Population Adjusted Doses (aPAD) or chronic Population Adjusted Doses (cPAD). Risk reduction measures are typically invoked in cases where EPA judges that risks at the 99.9th level are excessive. Our contractor analyzed the distribution of both food consumption data and pesticide residue levels to produce an estimate of dietary risk that reflects the upper end of the pesticide exposure distribution. By "upper-end," we mean between the 95th and 99.9th levels of exposure. We approximate the 99.9th level of exposure to a given pesticide by combining estimates of food 1 Some of the analysis work was conducted through a contract with Benbrook Consulting Services, Sandpoint, Idaho. 2 Hereafter in this report, the 95th, 99th, or 99.9th percentile level of a distribution of values are referred to as just the 95th, 99th, or 99.9th level. S-1 ------- consumption that reflect approximately the 95th level of consumption, with an estimate of pesticide residue levels at about the 99th level of the distribution of positive values. Dietary Risk Index Our basic unit of measure used to track pesticide dietary risks is called the Dietary Risk Index (DRI). DRI values or scores are calculated for each pesticide-food combination covered in the annual testing carried out by the PDP. For a given food and year, DRI values for each pesticide found in the food are added together, to form an aggregate, food-level DRI score. Single-food and aggregate DRI scores are calculated for three sets of residue data: food grown, harvested, and processed in the United States (domestic production); residue in food that is imported into the United States; and all samples tested by the PDP in a given year (domestic plus imported samples, plus samples of unknown origin). Trends over time in aggregate food-level DRI scores provide insights into changes in overall risk levels and the relative share of total risk accounted for by residues in imports versus domestic foods. Our contractor calculated DRIs using assumptions, methods, and data as close as possible to those called for in FQPA science policies and adhered to in recent EPA dietary risk assessments. The DRI is calculated from two variables: • "Percent Positive" - How frequently a pesticide residue is found in a food; and • "Chronic Risk Share" - The level of risk associated with the residues of a pesticide found in a food, taking into account the pesticide's toxicity, the amount of food typically eaten by children, and the mean of the residues found in positive samples. The basic formula to calculate the DRI score for a given pesticide-food combination is DRI = (Percent Positive) x (Chronic Risk Share) The "Percent Positive" variable is calculated from PDP data and equals the number of positive samples tested in a given year, divided by the total number of samples. For each pesticide-food combination, there are up to three "Percent Positive" values: one representing the results for domestic samples, one for imports, and one for all samples combined.3 DRI values can be calculated based on acute Reference Doses (aRfD) and aPAD, as well as chronic Reference Doses (cRfD) and cPAD. The analysis of dietary risk trends in this report is based on chronic risks, because EPA has not established aRfD for a majority of pesticides. 3 Each year, the PDP tests a few samples of "unknown origin." These samples are excluded from both the domestic production and imports analyses, but are included in the "combined" analyses. This is why the sum of samples, and the sum of positives in the domestic plus imported samples, sometimes is less than the number of samples and positives in the combined analysis. S-2 ------- Chronic Risk Share The Chronic Risk Share (CRS) is a new analytical concept developed for this report. It is designed to help answer a simple question - "How risky are the residues found in a given food?" The CRS is a measure of the degree to which the residues found in the food, as reported in PDP results, fill up the pesticide's "risk cup" for a person of known weight.4 A CRS for a given pesticide-food combination is calculated from two components. The first is called the "Projected 99th Residue Level" (or PRL99), which is an estimate of the residue level found in the food at about the 99th percentile level of the distribution of residues ranked from the lowest to the highest. The second component used to calculate the CRS is the pesticide's single-food chronic Reference Concentration (cRfC). Four variables are needed to calculate a single-food cRfC for a person of known weight - the average amount of food consumed by the person; the weight of the person; the toxicity of the pesticide; and the magnitude of exposures from other foods, beverages, or pesticide uses around the home, schools, or other residential settings. A single- food cRfC is an estimate of the concentration of a pesticide that can be present in a serving of a given food, without exceeding the person's chronic PAD. The CRS for a given pesticide-food combination is calculated as follows CRS = (PRLqq) (Single-Food cRfC) In cases where the PRL99 exceeds the applicable single-food cRfC, the value of the CRS will be greater than one. In such cases, a small portion of the people consuming the food in a given day is likely to receive a dose of the pesticide above the level EPA regards as acceptable. The smaller the value of the CRS, the less worrisome the dietary risks stemming from the residues present in a given food. For example, if even the 99th percentile residue level accounts for only one-tenth of the pesticide's single food cRfC, there is little reason for concern from dietary exposure, especially when EPA is confident that it has fully accounted for all other routes of exposure in setting the single-food cRfC. 4 EPA introduced the "risk cup" concept to help explain the impact of the FQPA on the allowable level of exposure to a given pesticide through all routes of exposure, taking into account whether any other pesticides pose risks through a common mechanism of action. The "risk cup" is a graphical representation of the acceptable amount of exposure to a given pesticide for a person of known weight. The size of the risk cup is typically reported in milligrams of pesticide per day, and is based on the pesticide's inherent toxicity and the average weight of a child exposed to the pesticide. S-3 ------- Projected 99th Residue Levels The Consumers Union5 studied the distribution of residue levels found in 53 food-pesticide combinations based on PDP testing in 1997 (Consumers Union/Natural Resources Defense Council, 1999).6 For each pesticide food combination, the Consumers Union identified the minimum and maximum residue, the mean of the positives, and the residue level at the 99.9th, 99th, and 95th levels of the distribution. Ratios were also calculated showing the difference between the 99.9th level and the mean, the 99th level and the mean, and the 95th level and the mean. The average difference between the 99.9th residue and the mean of the positives across the 53 food-pesticide combinations was 8.5. The difference between the 99th and the mean residue level was 6.1. To estimate dietary risks close to the 99.9th level chosen by EPA as the threshold of regulation, we selected a value of 7 as the average difference between the mean residue level and the PRL99 value. Accordingly, the formula for estimating the PRL99 level for a given pesticide-food combination is: PRL99 = (Mean residue) x 7 Single-Food Chronic Reference Concentrations The single-food chronic Reference Concentration, or cRfCsf, can be thought of as an initial estimate of the maximum level of a pesticide that can be present in a given food without violating the FQPA's basic "reasonable certainty of no harm" standard. This key concept is useful both in tracking changes in pesticide dietary risks and in setting the maximum levels for pesticide tolerances in food as eaten. A cRfCsf for a given pesticide will change as a function of the weight of a child and the amount (weight) of a specific food the child consumes during a day. In carrying out an analysis of changes over time in pesticide dietary risks, the assumptions used to set cRfCsf levels are less important than using consistent assumptions across all foods. This is because the goal is to identify changes in relative risk levels over time and across foods and pesticides, rather than estimating risk levels at a specific point in time, for comparison to some quantitative standard. The formula to calculate a cRfC for all foods and routes of exposure is: cRfC (mg/kg) x Serving Size Foody (grams/day) = Weight of Child (kg) x cPAD for Pesticidex (mg/kg/day) 5 Consumers Union is an independent nonprofit organization with a mission "to work for a fair, just, and safe marketplace for all consumers and to empower consumers to protect themselves." The organization accepts no outside advertising and no free test samples. Consumers Union supports itself through the sale of information products and services, individual contributions, and a few noncommercial grants. It has published a long list of FQPA-relevant reports since 1996. 6 For a complete discussion of the Consumers Union/Natural Resources Defense Council analysis of the distribution of PDP residue levels, see the comments submitted to EPA on June 6, 1999, on the Science Policy Paper, "Choosing a Percentile of Acute Dietary Exposure as a Threshold of Regulatory Concern" (Consumers Union, 1999; posted at http://www.ecologicipm.com/999 comments.pdf). S-4 ------- This equation can be solved for cRfC by converting the grams of food on the left side of the equation to kilograms of food, and then dividing by "Serving Size Foodx: cRfC (mg/kg) = Weight of Child (kg) x cPAD (mg/kg/dav) Serving Size Foody (kg/day) The weight of the child used in this report to calculate cRfC values is 16 kilograms. This is the value corresponding to the 50th percentile of growth for a 4-year-old male, as reported on the Centers for Disease Control and Prevention Growth Chart. EPA sets pesticide cPADs based on animal experiments, after applying a set of safety factors to the "No Observable Adverse Effect Level" for the most sensitive biological impact considered relevant in assessing the pesticide's toxicity. Our contractor analyzed the distribution of food consumption values for children's foods based on USDA's "Continuing Survey of Food Intake by Individuals" (commonly referred to as the CSFII survey). The survey contains 5,372 valid eating days for l-to-5-year-old children. For each of these foods, the most common amount reported was either exactly equal to, or close to what the USDA reports as, the typical serving size in its National Nutrient Database for Standard Reference (Release 17).7 In general, the 95th level of consumption is at least two-thirds larger than the typical serving size reported by USDA. We have listed in the table below foods and their portion sizes used in our calculations. When we lacked data from USDA on the portion size at the 95th level of consumption, we multiplied the typical serving size by 1.667. Estimated 95tn Percentile Food Consumption Levels Used in Calculating Chronic Risk Shares and Dietary Risk Index Values Approximate Equivalent Serving at 95th Level of Consumption 95th Percentile (grams) Apple Juice 3 cups 744 Apples 1 large 212 Bananas 1 large 136 Broccoli, raw 1.667 stalks 247 Cantaloupe, fresh 1.33 cups 223 Carrots 1 large 72 Celery 2.5 large stalks 183 Cucumbers 1-8" 301 Grapes, red or green, raw 1.33 cups 210 Green Beans, fresh, raw 1.667 cups 183 Lettuce 2 cups 148 Oranges 1 large 184 Peaches 1 large 157 Pears 2 medium 332 Potatoes, baked with skin 1 large 300 Spinach, raw 1.667 cups 50 Sweet Bell Peppers 1 medium 119 Tomatoes, raw 2 medium 247 7 Accessible at http://www.nal.usda.gov/fnic/foodcomp/Data/SR17/srl7.html S-5 ------- Children are typically exposed to a given pesticide through more than one food. Pesticides that appear in foods may also sometimes be present in fruit juices, other beverages, and drinking water. In addition, the FQPA directed EPA to take into account residential, schoolyard, and all other potential routes of exposure to a pesticide in setting and reviewing tolerances. For these reasons, if a single food accounts for the total allowed exposure to a pesticide on a given day, the child would almost certainly be overexposed because of residues in other foods and drinks, and possibly residential exposures. In its October 13, 2000 comments to EPA on chlorpyrifos risk mitigation (Consumers Union, 2000),8 the Consumers Union recommended that EPA not allow any single food use of a pesticide, including chlorpyrifos, to account for more than 10 percent of the pesticide's risk cup, at least not until EPA completed its cumulative risk assessment of the organophosphates and had taken all regulatory actions needed to meet the FQPA's "reasonable certainty of no harm" standard. We have incorporated this recommendation into the calculation of "Single-Food Chronic Reference Concentrations," which equal the total cRfC for a pesticide divided by 10.9 g Accessible at http://www.ecologic-ipm.com/Chlorpyrifos comments 2000.1x11' 9 This assumption likely biases single-food cRfC values upward for pesticides that appear routinely as residues in more than 10 foods. Likewise, cRfCsfvalues for pesticides found in just a few foods, and also not used in residential settings, are probably biased downward. These sources of bias do not impact the validity of results when single food cRfC values are used consistently in projecting relative dietary risk levels over time and across foods and pesticides. S-6 ------- Aggregate Dietary Risk Index Values for Selected Foods Aggregate Dietary Ri Domestically Grown sk Index Values for Se Plus Imported Foods: lected Foods Grown Domestically PDP Test Results for 1994-2003, \A Imported to the Ur fith Interpolated Va lited States, and for ues for Un-sampled Years Crop 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Apple Juice Domestic 00 © CO o 18.08 18.08 32.73 26.67 21.25 15.83 10.41 4.99 4.99 Import 97.44 97.44 97.44 67.15 28.27 22.45 16.64 10.82 5.00 5.00 Apples Domestic 316.00 295.48 359.35 299.63 239.90 180.18 144.33 144.15 43.65 43.65 Import 236.67 289.35 510.83 581.63 652.43 723.23 410.10 96.97 30.21 30.21 Broccoli Domestic 8.83 11.58 14.33 17.09 19.84 22.59 25.35 28.10 12.91 12.91 Import 97.05 98.82 100.59 102.36 104.13 105.89 107.66 109.43 62.29 62.29 Cantaloupe Domestic 36.66 36.66 36.66 36.66 36.66 55.46 85.46 62.15 38.85 15.54 Import 56.23 56.23 56.23 56.23 56.23 100.55 118.69 89.42 60.15 30.87 Carrots Domestic 11.40 11.17 13.79 13.29 12.79 12.29 11.78 9.46 9.82 9.82 Import 37.34 38.26 28.92 38.10 47.27 56.45 65.62 25.08 30.26 30.26 Celery Domestic 91.60 90.43 89.26 88.09 86.91 85.74 84.57 83.40 104.49 104.49 Import 50.41 143.18 235.95 328.72 421.49 514.26 607.03 699.80 170.00 170.00 Cucumbers Domestic 343.42 343.42 343.42 343.42 343.42 343.42 236.64 157.04 77.45 92.75 Import 460.92 460.92 460.92 460.92 460.92 460.92 443.69 354.86 266.03 317.16 Grapes Domestic 404.46 177.58 128.55 101.78 75.00 48.22 21.44 18.54 18.54 18.54 Import 194.17 240.79 381.66 402.80 423.94 445.07 466.21 281.75 281.75 281.75 Green Beans Domestic 378.04 293.97 311.45 328.94 346.42 363.91 381.39 330.10 330.10 330.10 Import 80.65 40.49 48.71 56.92 65.13 73.34 81.55 93.31 93.31 93.31 Lettuce Domestic 152.13 123.96 95.79 67.62 39.46 11.29 12.87 54.60 54.60 54.60 Import 90.38 90.38 90.38 90.38 90.38 90.38 325.58 925.79 925.79 925.79 Oranges Domestic 37.47 16.03 28.48 22.27 16.06 9.85 3.64 4.00 4.00 4.00 Import 2.26 2.26 2.08 1.91 1.73 1.55 1.37 1.65 1.65 1.65 111 The DRI values in red are interpolated or extrapolated from the values in years when PDP tested the food based on two assumptions: 1. The DRI value in the first year a food was tested was assigned to any earlier years back to 1994, and the DRI value in the last year the food was tested was assigned to any years after, up to and including 2003. 2. In cases with a gap between PDP samplings, we assumed that DRI values changed linearly during the time period when the food was not included in the program. S-7 ------- Aggregate Dietary Ri Domestically Grown sk Index Values for Se Plus Imported Foods: lected Foods Grown Domestically PDP Test Results for 1994-2003, \A Imported to the Ur fith Interpolated Va lited States, and for ues for Un-sampled Years Crop 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Peaches Domestic 766.07 942.95 836.73 687.27 537.80 388.34 238.88 89.42 54.06 54.06 Import 84.63 129.21 150.63 167.28 183.93 200.58 217.23 233.88 266.11 266.11 Potatoes Domestic 84.86 72.73 64.45 56.16 47.88 39.60 31.32 64.11 73.80 73.80 Import 6.00 12.53 14.41 16.28 18.16 20.03 21.91 23.78 22.45 22.45 Spinach Domestic 52.59 52.59 37.71 28.84 27.36 25.88 24.41 22.93 21.45 17.88 Import 27.47 27.47 83.69 16.21 13.57 10.93 8.29 5.65 3.01 9.08 Sweet bell peppers Domestic 243.25 243.25 243.25 243.25 243.25 243.25 180.80 164.49 148.18 132.02 Import 1,067.45 1,067.45 1,067.45 1,067.45 1,067.45 1,067.45 874.19 658.74 443.29 720.25 Tomatoes Domestic 175.92 175.92 175.92 189.91 123.37 103.36 94.62 85.88 77.14 68.40 Import 543.88 543.88 543.88 580.21 453.44 478.62 394.37 310.12 225.87 141.62 Total and Average Dietary Risk Index Values (With Values Interpolated for Un-sampled Years) TOTAL DIETARY RISK INDEX (DRI)* Domestic 3,120.76 2,905.78 2,797.21 2,556.94 2,222.81 1,954.64 1,593.34 1,328.78 1,074.01 1,037.53 Import 3,132.95 3,338.68 3,873.76 4,034.53 4,088.44 4,371.69 4,160.12 3,921.04 2,887.16 3,107.81 Average DRI Values Domestic 195.05 181.61 174.83 159.81 138.93 122.16 99.58 83.05 67.13 64.85 Import 195.81 208.67 242.11 252.16 255.53 273.23 260.01 245.06 180.45 194.24 Total and Average Dietary Risk Index Values (With Missing Values for Un-sampled Years) TOTAL DIETARY RISK INDEX (DRI)* Domestic 2,250.84 1,862.48 1,598.61 251.49 186.71 936.96 1,109.68 825.87 550.78 326.58 Import 786.92 780.38 1,797.05 663.57 537.94 2,921.15 2,376.90 2,491.43 1,298.64 1,218.99 Number of Foods Tested Domestic 10 8 8 3 3 6 10 10 10 5 Import 8 8 7 3 3 6 8 10 10 5 Average DRI Values Domestic 225.08 232.81 199.83 83.83 62.24 156.16 110.97 82.59 55.08 65.32 Import 98.37 97.55 256.72 221.19 179.31 486.86 297.11 249.14 129.86 243.80 S-8 ------- Impact of EPA Actions on Risk Driver Pesticide-Food Combinations from Domestic Commodities Impact of EPA Actions fror (Ranked by Percentage Cham on Risk Driver Pesticide-Food Combinations n Domestic Commodities ge in Dietary Risk Index Levels from the Pre-FQPA Period) Commodity Pesticide DRI Score Year Change in DRI Score Grapes Parathion methyl 0.0 2001 -100% 329.1 1994 Green Beans, Processed Parathion methyl 0.0 2004 -100% 22.6 1996 Peaches Parathion methyl 0.0 2004 -100% 799.4 1996 Pears Parathion methyl 0.0 2003 -100% 78.1 1997 Apples Parathion methyl 0.0 2004 -100% 52.0 1996 Tomatoes Chlorpyrifos 0.0 2004 -100% 36.8 1997 Wheat Flour Chlorpyrifos methyl 0.0 2004 -100% 149.2 1996 Spinach, Processed Parathion ethyl 0.6 1999 -99% 88.2 1998 Apples Chlorpyrifos 3.6 2002 -98% 207.3 1996 Strawberries Vinclozolin 4.4 2000 -93% 65.7 1998 Grapes Dicofol p,p' 12.3 2001 -85% 82.7 1996 Green Beans, Processed Methamidophos 15.2 2003 -83% 89.1 1996 Strawberries Dicofol p,p' 13.4 2000 -80% 67.3 1998 Tomatoes Methamidophos 34.9 2003 -76% 143.4 1996 Cucumbers Dieldrin 33.6 2003 -70% 111.3 1999 Sweet Bell Peppers Chlorpyrifos 20.7 2003 -68% 65.0 1999 Pears Azinphos methyl 19.6 2003 -67% 58.6 1997 S-9 ------- Impact of EPA Actions on Risk Driver Pesticide-Food Combinations from Domestic Commodities (Ranked by Percentage Cham ge in Dietary Risk Index Levels from the Pre-FQPA Period) Commodity Pesticide DRI Score Year Change in DRI Score Winter Squash Dieldrin 77.8 1999 -57% 179.3 1997 Cucumbers Methamidophos 13.4 2003 -55% 29.8 1999 Sweet Bell Peppers Methamidophos 60.9 2003 -49% 119.1 1999 Green Beans Endosulfans or 20.0 2001 -47% endosulfan sulfate 37.6 1995 Strawberries, Dicofol p,p' 22.8 2000 -37% Processed 36.1 1998 Winter Squash, Dieldrin 228.5 1999 -36% Processed 354.4 1997 Green Beans Dimethoate 20.6 2001 -31% 30.0 1994 Green Beans Acephate 55.5 2001 9% 51.0 1994 Green Beans Methamidophos 205.1 2001 23% 166.3 1995 Celery Acephate 36.5 2002 25% 29.2 1994 Strawberries, Vinclozolin 31.6 2000 77% Processed 17.9 1998 Potatoes Chlorpropham 61.1 2002 114% 28.5 1995 Peaches Dicofol p,p' 16.6 2001 1975% 0.8 1996 Impact of EPA Actions 1649.0 % The two columns to the right show how nine EPA actions reduced DRI scores in this table. Nine EPA actions reduced DRI scores by about 1650; EPA revoked the tolerances for 7 of the 30 risk pairs listed (6 parathion food uses and Parathions 1369.4 83% Chlorpyrifos 240.5 15% chlorpyrifos in tomatoes), substantially lowered the tolerance for chlorpyrifos in apples, and modestly lowered the tolerance for azinphos-methyl for pears. The actions taken on methyl or ethyl parathion resulted in an aggregate drop in DRI scores of about 1370, or 83 percent of the total impact triggered by tolerance revocations. The changes in chlorpyrifos tolerances reduced aggregate DRI scores by 240. Taken together, tolerance revocations and reductions imposed on eight uses of methyl and ethyl parathion and chlorpyrifos accounted for 98 percent of the total impact of EPA actions on the above set of information. Para+chlor 1610.0 98% S-10 ------- Impact of EPA Actions on Risk Driver Pesticide-Food Combinations from imported Sampies Impact of EPA Actions on Risk Driver Pesticide-Food Combinations from Imported Samples (Ranked by Percentage Change in Dietary Risk Index Levels from the Pre-FQPA Period) Commodity Pesticide DRI Score Year Change in DRI Score Green Beans, Processed Parathion methyl 0.0 2003 -100% 200.6 1997 Broccoli Mevinphos 0.0 2002 -100% 95.0 1994 Grapes Mevinphos 0.0 2001 -100% 34.5 1994 Apples Chlorpyrifos 13.9 2002 -97% 454.9 1996 Tomatoes Chlorpyrifos 68.8 2003 -85% 451.8 1996 Apple Juice Dimethoate 11.1 1998 -83% 65.0 1996 Sweet Bell Peppers Methamidophos 92.3 2003 -72% 327.8 1999 Winter Squash Dieldrin 1.1 1999 -68% 3.4 1997 Cucumbers Endosulfan I 15.0 2003 -55% 33.5 1999 Grapes Dimethoate 21.6 2001 -38% 35.1 1996 Pears Azinphos methyl 30.7 1999 -33% 45.8 1997 Cucumbers Methamidophos 179.8 2003 -32% 264.4 1999 Green Beans, Processed Methamidophos 27.6 2003 -10% 30.5 1997 Tomatoes Methamidophos 44.1 2003 -8% 48.0 1996 Sweet Bell Peppers Chlorpyrifos 586.6 2003 -2% 595.6 1999 Celery Acephate 16.6 2002 8% 15.4 1994 Peaches Dicofol p,p' 22.3 2002 18% 18.9 1996 S-11 ------- Impact of EPA Actions on Risk Driver Pesticide-Food Combinations from Imported Samples (Ranked by Percentage Change in Dietary Risk Index Levels from the Pre-FQPA Period) Commodity Pesticide DRI Score Year Change in DRI Score Celery Methamidophos 104.3 2002 225% 17.2 2001 Pears Dicofol p,p' 120.6 1999 323% 28.5 1998 Cantaloupe Methamidophos 92.7 2000 409% 18.2 1998 Strawberries Endosulfan I 71.3 1999 413% 13.9 1998 Potatoes Chlorpropham 14.5 2002 1971% 0.7 1995 Peaches Methamidophos 31.1 2002 10267 % 0.3 1996 Impact of EPA Actions 1039.7 % Parathions 200.6 19% Chlorpyrifos 824.0 79% Para+chlor 1024.6 99% S-12 ------- |