56Expert Panel Report Final With Graphics

8/9/2019 56Expert Panel Report Final With Graphics

1/20

E x p E r t p a n E l r E p o r t

a reie he Siee

he pei

Heh Efes

pesiie resies

F ree

Semes Me bIees Gs

Authors:

Penny Fenner-Crisp,

EPA, Retired

Carl L. Keen,

University of California, Davis

Jason Richardson,

Robert Wood Johnson Medical School

Rudy Richardson,

University of Michigan

Karl Rozman,University of Kansas

2010


2/20

ExEcutIvE SuMMary

This report describes the deliberations andconclusions of a scientic expert panelassembled to evaluate statements made by the

Environmental Working Group (EWG) and theOrganic Trade Association (OTA) regarding thepotential health effects of pesticide residues onfood and the nutritional quality of organically-grown food compared to food grown usingconventional agricultural methods. The panelwas commissioned by the Alliance for Foodand Farming, but the sponsor did not participatein the production of this report.

The EWG has recently assembled a list of 47

fruits and vegetables for which they haveanalyzed publicly-available data to determinethe number and magnitude of pesticide residuesdetected on these commodities. This listincludes a subgroup that EWG has termed thedirty dozen, asserting that these 12 foodscontain the highest levels and/or numbers ofpesticides relative to other commonly availableproduce in the United States, and implying thatthere are known to be adverse health effectsassociated with consuming these foods that aredue to the presence of these pesticide residues.

For example, the EWG states that The growingconsensus among scientists is that small dosesof pesticides and other chemicals can causelasting damage to human health, especiallyduring fetal development and early childhood.Small is not dened. The OTA has madesimilar statements with a focus on the potentialnegative effects to children, although it hasapparently not conducted any relevantindependent analysis of exposure or toxicitydata or the epidemiology literature.

The panel has reviewed the materials preparedby the EWG and the OTA and came to thefollowing conclusions:

The EWGs list may reect a relatively accurateordering of the listing of the 47 commoditiesfrom the highest to lowest levels/numbersof pesticide residues. However, the list ismisleading to consumers in that it is based

only upon exposure data while remainingsilent about available information on theassessment of the toxicity of pesticides

presented in the diet, and, as such, does notprovide a basis to assess risk. There also is noacknowledgment of the fact that the datashow that the residue levels detected are,with very rare exception, below or, morelikely, well below, the legal limits establishedonly after calculating the potential total non-occupational exposure that an individualmight experience to a pesticide approved foruse on an agricultural commodity.Furthermore, it is disconcerting that EWGdoes not describe its methodology in sufcient

detail so that others can duplicate theiranalysis and independently judge itscredibility, particularly given the widespreadpress coverage that its Shoppers Guide toPesticides has received.

The Panel does not agree with EWGs assertionthat there is a growing consensus amongscientists that the amount of pesticideresidues currently found on food constitutes asignicant public health issue. While there

will always be some uncertainty associatedwith evaluating the possibility of small healthrisks, the available scientic data do notindicate that this source constitutes asignicant risk.

The U.S. EPAs current process for evaluatingthe potential risks of pesticides on food isrigorous, and health-protective. The EPAstesting requirements for pesticides used onfood are more extensive than for chemicals inany other use category, and include testingtargeted specically to assess the potentialrisks to fetuses, infants, and children.

The currently-available scientic data do notprovide a convincing argument to concludethat there is a signicant difference betweenthe nutritional quality of organically grownfood and food grown with conventionalagricultural methods.

2


3/20


4/20

BackGround

EWG recently distributed an updated ShoppersGuide to Pesticides that lists 47 fruits andvegetables of which the top 12 commodities

were shown to have the highest detection rates/numbers of pesticide residues (the dirtydozen). The Guide also includes the Clean15, a subset of the 47 commodities whichwere shown to have the lowest levels/numbersof pesticide residues. The Guide is available insupermarkets across the country and also canbe downloaded from an EWG-afliated website(www.foodnews.org).

The Guide includes a brief description of the

methodology used to construct the list. A relatedEWG website contains slightly more informationon the basis for the list, including a list ofpublished references that were presumablyused in its development. The site contains tworelevant documents including a Methodologypiece that presents a cursory description ofEWGs methods for selecting the 47 commoditiesand a How to Reduce Exposure section thatincludes additional information about healthimpacts, including a list of citations that EWGalleges supports its claims.

EWGs dirty dozen list is as follows (startingwith the worst):

1. Peach

2. Apple

3. Bell pepper

4. Celery

5. Nectarine

6. Strawberries

7. Cherries8. Kale

9. Lettuce

10. Grapes (imported)

11. Carrot

12. Pear

EWGs Clean 15 includes (starting withthe best):

1. Onion

2. Avocado

3. Sweet corn

4. Pineapple

5. Mango

6. Asparagus

7. Sweet peas

8. Kiwi

9. Cabbage

10. Eggplant11. Papaya

12. Watermelon

13. Broccoli

14. Tomato

15. Sweet potato

EWG assembled the list by analyzing databasesof pesticide residue measurements collected bythe U.S. Department of Agriculture (USDA) in

its Pesticide Data Program (PDP) and theRegulatory Monitoring Program and Total DietStudy of FDAs Center for Food Safety andApplied Nutrition.

Within the EWGs report, the discussion of theputative health effects of pesticide residues isvery limited, and thus difcult to criticallyevaluate. The only reference to this topic is theintroductory paragraph in the Shoppers Guidewhere EWG states:

The growing consensus among scientistsis that small doses of pesticides and otherchemicals can cause lasting damage tohuman health, especially during fetaldevelopment and early childhood. Scientistsnow know enough about the long-termconsequences of ingesting these powerfulchemicals to advise that we minimize ourconsumption of pesticides.

4


5/20

The above statement does not include anycitations, thus complicating a direct evaluationof its relevance with respect to the amounts ofresidues that have been reported to be presenton the foods listed on the dirty dozen list.

Another statement in the How to reduceexposure piece states that:

Even in the face of a growing body ofevidence, pesticide manufacturers continueto defend their products, claiming that theamounts of pesticides on produce are notsufcient to elicit safety concerns. Yet,such statements are often made in theabsence of actual data, since most safety

tests done for regulatory agencies are notdesigned to discover whether low doseexposures to mixtures of pesticides andother toxic chemicals are safe, particularlyduring critical periods of development. Ingeneral, the government demands, andcompanies conduct, high dose studiesdesigned to nd gross, obvious toxiceffects. In the absence of the appropriatetests at lower doses, pesticide and chemicalmanufacturers claim safety since the fulleffects of exposure to these mixtures ofchemicals have not been conclusivelydemonstrated (or even studied).

The most relevant points in this section are thecontentions that studies do not exist on lowdoses of pesticides and pesticide mixtures, bothof which are addressed later.

The How to Reduce Exposure piece alsoraises issues associated with increasedvulnerability of children and criticism of theEnvironmental Protection Agencys (EPAs)regulation of pesticides.

Similar to EWG, the Organic Trade Association(OTA) has made statements about the healtheffects of pesticide residues. The OTA focuseson the potential effects of these residues onchildren1:

In the past decade, research and analysishas shown that children may be muchmore at risk than adults for pesticideexposure, and may suffer greater harm tohealth and development from exposure.Yet standards for safety and tolerance limitsfor these chemicals rarely include adequateconsideration of risks to children.

Recent laws now mandate factoring in

these risks and re-evaluating safety limits,but the wheels of re-evaluation have turnedvery slowly. [Note: OTA infers that this taskhas not been completed. However, it is inerror here. The re-evaluation of existingtolerances mandated by FQPA in 1996was completed in 2008]. Organic foods,therefore, may be especially important tomore fully protect children from the risksof exposure, even when pesticide levels infoods are within existing legal limits.

Why are children at greater risk? First, theyingest more food and water per pound ofbody weight than adults, so any exposureis greater in proportion to their size.Second, these chemicals may be moreharmful to developing organs and bodilysystems, including neurological andreproductive systems, than they are tomature bodies.

In a study published in May 2002 in FoodAdditives and Contaminants , organic foodswere shown to have signicantly lowerpesticide residues than conventionally

5

1 See http://organicitsworthit.com/environment.html.


6/20

grown foods (for a number of reasons, suchas persistent residues in soil that last formany years, some organic foods may stillshow residue).

Other studies show the environmentalbenets of organic agriculture to air, soiland water, lowering the total toxic burdento our ecosystems. As demand for organicfoods continues to grow, more farmers arelikely to view organic methods as a viableand marketable option, helping to stabilizesupply and price.

It adds up to an evolving landscape thatincreasingly allows for--and makes a

compelling and credible case for--includingorganic foods in childrens diets wheneverpossible. As concerned parents, teachers,administrators and foodservice professionalscreate and insist on innovation and reformin school lunch programs, organic foodsmake sense as part of the picture.

Regrettably, citations to scientic studies werenot provided in the above to support these

statements, complicating their criticalevaluation.

OTA has an additional document on pesticideexposures and children that focuses on studiesthat nd lower pesticide exposures for thosethat have organic diets and cites several studiesthat conclude that there health effects associatedwith pesticide use for farm workers.

cHarGE to tHE panEl

The panel was asked to address the followingissues:

1. The basis for the EWG ranking of thecommodities by pesticide residue levels/numbers to come up with the list of 47 fruitsand vegetables, including the dirty dozen.

2. The link between pesticide residues on fruitsand vegetables, and health effects, including:

a. Scientic evidence linking pesticideresidues and health effects

b. The adequacy of the U.S. regulatory systemfor protecting against harmful levels ofpesticide residues, including effects toinfants and children.

3. The evidence that organic foods have agreater nutritional quality than conventionally-grown foods.

cHarGE QuEStIon #1 IS tHE BaSIS

For SElEctInG tHE dIrty dozEn

ScIEntIFIcally Sound?

EWG briey describes its methodology forselecting the list of 47 fruits and vegetables,including the dirty dozen on theMethodology portion of its Shoppers Guideto Pesticides webpage. Data from the U.S.Department of Agriculture (USDA) PesticideData Program (PDP) and the Food and DrugAdministrations (FDA) Pesticide Regulatory

Monitoring and Total Diet Study Programs wereused as the basis for characterizing the numbersand levels of residues of pesticides on thecommodities. EWG focused on the 47 fruitsand vegetables that were reported eaten on atleast one tenth of one percent of all eatingdays identied in the 1994-1996 USDA foodconsumption survey and with a minimum of100 pesticide test results from the years 2000 to2007. EWG considered six measures ofcontamination on commodities:

1. Percent of samples tested with detectablepesticides

2. Percent of the samples with two or morepesticides

3. Average number of pesticides found ona sample

4. Average amount of all pesticides found

6


7/20


8/20

The growing consensus among scientistsis that small doses of pesticides and otherchemicals can cause lasting damage tohuman health, especially during fetaldevelopment and early childhood.

Scientists now know enough about thelong-term consequences of ingesting thesepowerful chemicals to advise that weminimize our consumption of pesticides.(EWG, undated)

In addition, EWG argues that toxicity testingrequired by EPA for pesticides is not designedto discover whether low dose exposures tomixtures of pesticides and other toxic chemicalsare safe, particularly during critical periods of

development. (EWG, undated)

The Organic Trade Association (OTA) focuseson potential effects for children and argues thatEPAs toxicity testing requirements areinadequate: Yet standards for safety andtolerance limits for these chemicals rarelyinclude adequate consideration of risks tochildren.3

Gee Siee pesiie resie

Heh Sies

Little published research directly addresses thepotential health effects of exposures to pesticideresidues in the diet. For example, epidemiologicstudies that compare populations with differentlevels of pesticide dietary exposures are lacking.The vast majority of studies to date that haveexamined the potential for health effectsresulting from pesticide exposure in children

are in populations with higher (and, primarily,non-dietary) exposures than the generalpopulation, including children of farm workersand pesticide applicators (Arcrury et al., 2007;Eskenazi et al., 2004), as well as childrenexposed through repeated indoor pesticide

application (Berkowitz et al., 2004). Of thesestudies, most have focused on theorganophosphate pesticides (e.g. chlorpyrifosand diazinon) and found that the levels thatthese populations were exposed to were much

higher than the general population. Based ondata from NHANES, the median level of theprimary metabolite of the pesticide chlorpyrifos,TCP, in the urine is 1.7 g/L, whereas medianlevels of TCP in the more highly-exposedpopulations are 45%, 94%, and 341% greaterthan the NHANES values (Arcrury et al., 2007;Eskenazi et al., 2004; Berkowitz et al., 2004).This comparison suggests that the predominantsources of exposure in these studies are fromnon-dietary sources.

The lack of published literature on health effectsarising directly from pesticide residues in foodwould seem to be evidenced in the fact thatneither EWG nor OTA cite a single study thatspecically examines exposure via this pathway.Most of the studies that EWG and OTA citeaddress exposures as a consequence ofoccupational activities or in environments at/near application sites (e.g., Andersen et al.,2008; Garry et al., 2002; Hoppin et al., 2006).These scenarios generally result in exposures

substantially greater than dietary exposures. Forexample, in EPAs chlorpyrifos risk assessment,the Agency estimates that the short-term dermalexposure for an aerial applicator to be 50 g/kg/day with an absorbed dose of 1.5 g/kg/day,assuming a 3% dermal absorption (EPA, 2006).The estimated inhalation exposure is 0.7 g/kg/day for a total dose estimate of 2.2 g/kg/day.By comparison, the estimated chronic dietaryexposure is 0.0008 g/kg/day and the estimatedacute dietary exposure is 0.02 g/kg/day. Thus,the estimated occupational exposure estimate

is between 100-3000 times higher than theestimated dietary exposure. Given that EPAuses the 99.9th percentile for acute dietaryexposure estimates and the 50th percentile forchronic dietary and occupational exposureestimates, the higher end of the range (3000) islikely the more accurate.

8

3 See http://organicitsworthit.com/environment.html.


9/20

Only one study cited by EWG is centered ondietary exposure (Petersen et al., 2008), but itfocuses on polychlorinated biphenyls (PCBs)and methyl mercury (neither of which arepesticides) and only secondarily addresses

occupational exposure to pesticides. Otherstudies cited by EWG focus on non-pesticidessuch as PCBs, phthalates and dioxins (Lundqvistet al., 2006; Stewart et al., 2008; Swan et al.,2005).

There is a substantial literature on the healthbenets of consuming fruits and vegetables.Numerous published studies show that theconsumption of fruit and vegetable-rich diets isassociated with a reduced risk for high blood

pressure; reduced risk of heart disease, stroke,and probably some cancers; and a lower risk ofocular and digestive problems4 (e.g., Law et al.,1998; Liu and Russell, 2008; Joshipura et al.,1999; Appel et al., 1997).

Individuals who consume large amounts offruits and vegetables likely have higher dietaryconsumption of pesticides, compared toindividuals with lower fruit and vegetableconsumption5. Of course, the research showingthe positive effects of fruit and vegetableconsumption does not shed much light on thequestion of whether or not the presence of lowlevels of pesticide residues may detract from, orhave no impact on, the benecial effects ofconsuming these foods. However, it stronglysupports the hypothesis that some of the allegedadverse effects of dietary consumption of lowlevel pesticide residues are not of the samescale as the benecial effects of consumingfruits and vegetables; otherwise, the adverseeffects from dietary pesticide consumption

would be evident in these studies.

Epas reg pess

While there is little scientic literature thatdirectly addresses potential adverse effects frompesticide exposures in the diet, the safety of the

U.S. food supply with respect to pesticideresidues can be evaluated by examining EPAsregulatory process.

Some of the most important points about EPAsregulatory process include:

EPA requires more toxicity testing forpesticides used on food than any use categoryof chemicals.

The development of toxicity reference levelsfor pesticides representing a reasonablecertainty of no harm includes theincorporation of uncertainty factors that serveto achieve this regulatory standard. Typically,assessments include at least a 10-folduncertainty factor for extrapolating fromanimals to humans, and a 10-fold factor forintraspecies variability, unless empirical dataare available to show a different factor betterreects the data at hand. Furthermore, EPA,

when establishing tolerances (the legal limitson foods) must include an additional 10-foldsafety factor for infants, children or fetusesunless there is convincing evidence that adifferent factor is appropriate.

As a default, cancer risk is evaluated using alinear, no-threshold model and a 1 in amillion acceptable risk level, unless theavailable data support the use of a margin-of-exposure approach.

For acute exposures, EPA bases the assessmenton the 99.9th percentile of exposure fordifferent subpopulations, which is greaterthan the percentiles typically used in riskassessments in other EPA programs.

9

4 http://www.hsph.harvard.edu/nutritionsource/what-should-you-eat/vegetables-and-fruits/.

5 It is true that an organic diet will lead to lower pesticide residue consumption. However, only a relatively small fraction of the populationconsumes only organic food and many of the studies showing the benets of fruits and vegetables contain subjects for which organic dietswere not available for most of their lives.


10/20

EPA is obligated to assess the aggregate risk toa single pesticide from all dietary and non-occupational exposures when decidingwhether or not to approve a new or continueduse on a single commodity.

EPA also must evaluate the combined riskassociated with pesticides and othersubstances to which the general populationmay be exposed that have a commonmechanism of toxicity using cumulative riskassessment methods. To date, the members ofgroups of organophosphate, N-methylcarbamates, chlorotriazine, and chloracetanilidepesticides have been assessed. A large groupof synthetic pyrethroid insecticides are

currently undergoing evaluation.

In accordance with the mandates of the FoodQuality Protection Act of 1996, EPAs updatedrisk assessments have resulted in the reductionof use rates, numbers of allowable uses, andthe cancellation of all registrations of manychemicals and product formulations.

The U.S. Department of Agriculture (USDA)manages a monitoring program which measures

levels of pesticide residues on a wide variety offoods. The Pesticide Data Program (PDP) dataindicate that pesticide residues measured ondomestic and/or imported commodities rarelyexceed EPA tolerances, and, generally, are oneor more orders of magnitude below the legallimit. In 2007, residues exceeding the EPAtolerance were detected in only 0.4% of 11,683samples (USDA, 2008). While it would bedesirable to further limit the already smallnumber of samples that have residues exceedingtolerances, it is important to note that the

toxicity of a pesticide does not factor intoestablishing a tolerance, and the tolerance levelrepresents an exposure that is often substantiallyless than levels shown to cause effects in animaltesting.

Smm csis

chge Qesi #2

The Panels summary conclusions include:

1. Pesticide residues on food represent a smallexposure compared to occupationalexposure. There are no studies that specicallylink pesticide residues in the diet with healtheffects. Those epidemiologic studies thatposit a link to health effects evaluatepopulations living in primarily agriculturalenvironments and who are also exposed viaother pathways. However, even these studiesare insufcient to establish causalrelationships. The exposures of these subjects

are primarily from pathways in addition tofood, with these pathways accounting formuch higher levels of exposure. These studiesare not capable of assessing any contributionthat pesticide residues in the dietmay maketo the risk of exposure to these substances.

2. EPA has adopted a public health protectiveapproach to ensure a reasonable certaintyof no harm (the legal standard mandated inFQPA) from consuming pesticide residueson food. It incorporates the most sophisticated,

data-rich set of risk assessment methods thatEPA conducts. Contrary to OTAs assertion,the process explicitly considers infants,children and pregnant women and has anadded layer of protection for thesesubpopulations. While there will always besome uncertainty associated with evaluatingthe possibility of small health risks, theavailable scientic evidence shows that EPAsprocess is appropriately and adequatelyhealth-protective.

3. EWG states that there is a growing consensusamong scientists that small doses ofpesticides and other chemicals can causelasting damage to human health, especiallyduring fetal development and earlychildhood. If small doses is understood tomean the doses one receives from pesticideresidues in food, this statement is not supportedby the existing scientic evidence.

10


11/20

4. The EWG has provided a list of scienticpublications to justify their claims abouthealth effects of pesticide residues. None ofthe papers cited differentiated dietaryexposures from other pathways. Therefore,

none of the studies is sufcient to draw aconclusion that there are adverse healtheffects associated with pesticide residues onfood.

5. EWG states Scientists now know enoughabout the long-term consequences ofingesting these powerful chemicals to advisethat we minimize our consumption ofpesticides. The Panel agrees that pesticideintake should be limited; it is the opinion of

the Panel that EPA does a sound job inlimiting it to levels meeting the reasonablecertainty of no harm FQPA standard.

6. EWG implies that toxicity tests are inadequate.In contrast to this idea, the Panel notes thatEPA requires more data for pesticides residueson food than for chemical in other usecategories. Contrary to EWGs assertion,these studies must include at least one dosethat shows no effects. If the study results donot reveal a no-effect level, then either thestudy must be repeated until a no-effect levelis identied or have an additional uncertaintyfactor applied to the lowest dose showingminimal effects, yielding a surrogate no-effect level. There is also a requirement fordevelopmental neurotoxicity testing, designedto assess the potential for neurological effectson developing fetuses and children, for thosepesticides known or suspected of possessingneurotoxic potential.

cHarGE QuEStIon #3 IS tHErE

a dIFFErEncE In tHE nutrItIonal

QualIty oF orGanIcally-Grown

Food coMparEd to Food Grown

uSInG convEntIonalaGrIculturE?

There is a perception among many consumersthat organically-grown food is nutritionallysuperior in some respects to food grown withconventional agriculture. Two hypotheses havebeen put forward to explain the potentialdifferences. One hypothesis is thatconventionally-grown plants have morenitrogen available to them through the use of

synthetic fertilizers. As a consequence, theresources of the plants are diverted towardssupporting growth resulting in a decrease in theproduction of plant secondary metabolites suchas organic acids, polyphenolics, chlorophyll,and amino acids, all of which may have somenutritional benet (Winters and Davis, 2006).Another hypothesis is that organic productionmethods lead to greater stresses on plants. Astressed plant then may expend more resourcesin the synthesis of its own chemical defensemechanisms, which, in turn, may yield

substances which would not have positivenutritional effects (Winters and Davis, 2006).

Generally, controlled studies have shown mixedresults. Some support the conclusion thatorganic production methods lead to increasesin nutrients. Other studies show no demonstrabledifferences. A recent analysis conducted by theLondon School of Hygiene & Tropical Medicineprovides a comprehensive review of theavailable literature (Dangour et al., 2009). Theauthors identied 46 studies with sufcientdocumentation and quality upon which theyperformed a systematic review. Elevennutritional categories were evaluated. Thenitrogen content of conventionally-grownplants was higher, and the phosphorus andtitratable acidity levels were higher fororganically-grown plants. These differenceswere considered biologically plausible due to

11


12/20

differences in fertilizer use (nitrogen andphosphorus) and ripeness at harvest (titratableacidity). There was no difference for theremaining eight categories, including some keyones, including Vitamin C, phenolic compounds,

magnesium, calcium, potassium, zinc, totalsoluble solids, and copper. The authorsconcluded that:

The current analysis suggests that a smallnumber of differences in nutrient contentexist between organically and conventionallyproduced foodstuffs and that, whereasthese differences in content are biologicallyplausible, they are unlikely to be of publichealth relevance.

The authors encourage more research in this area.

The Scientic Status Summary on OrganicFoods from the Institute for Food Technologists(IFT) echoes the conclusions of the Londonreview (Winters and Davis, 2006). The IFTSummary discusses a variety of issuessurrounding organic foods, including: (1) levelsof pesticides, (2) nutritional value, (3) naturallyoccurring toxins, and (4) microbiological safety,

and includes a summary of a number of keystudies comparing organic and conventionalfoods with respect to nutrient levels.

The IFT Summary states:

In some cases, organic foods may havehigher levels of plant secondarymetabolites; this may be benecial withrespect to suspected antioxidants such aspolyphenolic compounds, but also may

be of potential health concern whenconsidering naturally occurring toxins.Some studies have suggested potentialincreased microbiological hazards fromorganic produce or animal products dueto prohibition of antimicrobial use, yetother studies have not reached the sameconclusion. Bacterial isolates from food

animals raised organically appear to showless resistance to antimicrobial agents thanthose food animals raised conventionally.

While many studies demonstrate thesequalitative differences between organicand conventional foods, it is premature toconclude that either food system is superiorto the other with respect to safety ornutritional composition. Pesticide residues,naturally occurring toxins, nitrates, andpolyphenolic compounds exert theirhealth risks or benets on a dose-relatedbasis, and data do not yet exist to ascertainwhether the difference in the levels of suchchemicals between organic foods and

conventional foods are of biologicalsignicance.

It is important to state that the nutrient levels innatural plants can vary for a wide variety ofreasons. It is plausible for plants grown underdifferent conditions, such as conventionalversus organic agriculture, to have differentnutritional qualities. However, there is noconvincing reason to believe that any oneproduction method is consistently superior inregard to nutrition. This is borne out by theavailable data which shows mixed resultsregarding systematic difference betweenfoodstuffs grown with conventional versusorganic agriculture.

It is also notable, as the IFT review details, thatthere is no convincing evidence of greatermicrobiological risk associated with organicfood, as some have suggested. Themicrobiological risk may be more related to thequality of the production method and the

prevention of contamination than from theparticular production method used.

12


13/20

rEFErEncES

Andersen, H.R.; Schmidt, I.M.; Grandjean, P.; Jensen, T.K.; Budtz-Jorgensen, E.; Kjrstad,M.B.; Blum, J.; Nielsen, J.B.; Skakkebk,

N.E.; Main, K.M. 2008. Impaired reproductivedevelopment in sons of women occupationallyexposed to pesticides during pregnancy,Environmental Health Perspectives, 116, 566-572.

Appel, L.J.; Moore, T.J.; Obarzanek, E., et al.1997. A clinical trial of the effects of dietarypatterns on blood pressure. DASH CollaborativeResearch Group. New England Journal ofMedicine. 336, 1117-1124.

Berkowitz, G.S.; Wetmur, J.G.; Birman-Deych,E.; Obel, J.; Lapinksi, R.H.; Godbold, J.H.;Holzman, I.R.; Wolff, M.S. 2004. In uteropesticide exposure, maternal paraoxonaseactivity, and head circumference. EnvironmentalHealth Perspectives, 112, 388-391.

Dangour, A.D., Dodhia, S.K., Hayter, A., Allen,E., Lock, K. and Uauy, R. 2009. Nutritionalquality of organic foods: a systematic review.

American Journal of Clinical Nutrition . 90, 680-685.

Environmental Working Group (undated)Materials on the Dirty Dozen available atwww.foodnews.org.

EPA. 2006. Reregistration eligibility decisionfor chlorpyrifos. Available at http://www.epa.gov/pesticides/reregistration/status.htm.

Eskanazi, B.; Harley, K.; Bradman, A.; Weltzien,E.; Jewell, N.P.; Barr, D.B.; Furlong, C.E.;Holland, N.T. 2004. Association of in uteroorganophosphate pesticide exposure and fetalgrowth and length of gestation in an agriculturalpopulation. Environmental Health Perspectives,112, 1116-1124.

Garry, V.F.; Harkins, M.E.; Erickson, L.L.; Long-Simpson, L.K.; Holland, S.E.; Burroughs, B.L.2002. Birth defects, season of conception, and

sex of children born to pesticide applicatorsliving in the Red River Valley on Minnesota,USA. Environmental Health Perspectives, 110Supplement 3, 441-449.

Hoppin, J.A.; Umbach, D.M.; London, S.J.;Lynch, C.F.; Alavanja, M.C.; Sandler, D.P. 2006.Pesticides and adult respiratory outcomes inthe Agricultural Health Study, Annals of theNew York Academy of Science, 1076, 343-354.

Joshipura, K.J.; Ascherio, A.; Manson, J.E., et al.1999. Fruit and vegetable intake in relation torisk of ischemic stroke.Journal of the AmericanMedical Association. 282, 1233-1239.

Klaassen, C. 2007. Casarett & Doulls toxicology:the basic science of poisons. McGraw-HillProfessional.

Law M.R.; Morris J.K. 1998. By how much does

fruit and vegetable consumption reduce therisk of ischaemic heart disease? European

Journal of Clinical Nutrition. 52, 549-556.

Liu C.; Russell R.M. 2008. Nutrition and gastriccancer risk: an update. Nutrition Review. 66,237-249.

Lundqvist, C.; Zuurbier, M.; Leijs, M.; Johansson,C.; Ceccatelli, S.; Saunders, M.; Schoeters, G.;ten Tusscher, G.; Koppe, J.G. 2006. Acta

Paediatr. Supplement. 95, 55-64.

Petersen, M.S.; Halling, J.; Bech, S.; Wermuth,L.; Weihe, P.; Nielsen, F.; Jrgensen, P.J.; Budtz-

Jrgensen, E.; Grandjean, P. 2008. Impact ofdietary exposure to food contaminants on therisk of Parkinsons disease. Neurotoxicology,29, 584-590.

13


14/20

Stewart, P.W.; Lonky, E.; Reihman, J.; Pagano, J.;Gump, B.B.; Darvill, T. 2008. The relationshipsbetween prenatal PCB exposure and intelligence(IQ) in 9-year old children. EnvironmentalHealth Perspectives, 116, 1416-1422.

Swan, S.H.; Main, K.M.; Liu, F.; Stewart, S.L.;Kruse, R.L.; Calafat, A.M.; Mao, C.S.; Redmon,

J.B.; Ternand, C.L.; Sullivan, S.; Teague, J.L. etal. 2005. Decrease in anogenital distanceamong male infants with prenatal phthalateexposure. Environmental Health Perspectives,113, 1056-1061.

USDA. 2008. Pesticide data program progressreport. Available at http://www.usda.gov/.

Winter, C.K. and Davis, S.F. 2006. ScienticStatus Summary for Organic Foods, Institute ofFood Technologists, Journal of Food Science.71, R117-R124.

14


15/20

d. pe Fee-cis

Dr. Fenner-Crisp served as the ExecutiveDirector of the ILSI Risk Science Institute (RSI)from December 2000 until August 2004,following a 22-year career at US EPA. Herduties at EPA included nearly 12 years servingin several capacities as the Senior ScienceAdvisor, Deputy Director and Director of theHealth Effects Division of the Ofce of PesticidePrograms. Earlier assignments included servingas the Director of the Health and EnvironmentalReview Division (HERD) of the Ofce of

Pollution Prevention and Toxics (OPPT) andSenior Toxicologist in the Health Effects Branchof the Ofce of Drinking Water (ODW). Sheplayed key roles in the development of manyEPA risk assessment policies and practicesprimarily related to human health and wasinvolved in the activities of several internationalorganizations as an expert on several WHOIPCS working groups, as a member of the WHOExpert Panel of the Joint Meeting on PesticideResidues for nine years and as the lead U.S.Delegate to several workgroups of the OECD

test guidelines program. In April, 2000, shereceived the Agencys highest award, theFitzhugh Green Award, for her contributions onbehalf of EPA to its international activities.

Dr. Fenner-Crisp received her Ph.D. inPharmacology from the University of TexasMedical Branch in Galveston and is a memberand former ofcer of several professionalscientic societies including of the Society ofToxicology and the Society for Risk Analysis.She has been a Diplomate of the AmericanBoard of Toxicology since 1984 and served onits Board of Directors from 2001-2005. Sheserved on EPAs Endocrine Disruptor MethodsValidation Subcommittee from 2001-2004 andthe Strategic Science Team of the AmericanChemistry Councils Long-range ResearchInitiative from 2002-2005. Currently, she is amember of the Board of Directors of theMidwest Center for Environmental Science and

Public Policy, the Drinking Water Committee ofEPAs Science Advisory Board and EPAs NationalPollution Prevention and Toxics AdvisoryCommittee. She also is a member of theNational Academies of Sciences expert groupcharged with conducting a review of the Workerand Public Health Activities Programadministered by the Department of Energy andthe Department of Health and HumanServices.

d. c l. kee

Dr. Carl L. Keen is the Mars Chair inDevelopmental Nutrition, Professor of Nutrition& Internal Medicine, and a Nutritionist in theAgricultural Experiment Station at the Universityof California at Davis. Dr. Keen received hisB.S. and Ph.D. degrees in Nutrition from theUniversity of California, Davis. Dr. Keensresearch group has four main areas of focus.The rst concerns the inuence of diet onembryonic and fetal development. A signicantproportion of birth defects are the consequenceof embryonic and fetal malnutrition. A thesis inthe laboratory is that the correction of suboptimalnutritional deciencies during earlydevelopment should result in a markedreduction in pregnancy complications. Thesecond research theme in the group is the studyof gene-nutrient interactions, with an emphasison how subtle changes in cell nutrientconcentrations can inuence the expression ofselect genes. The third major research theme inthe group is the study of how diet inuences

oxidant defense systems and cellular oxidativedamage. The fourth area of research in thelaboratory is on the effects of diet on thedevelopment and progression of vasculardisease. A current hypothesis in the laboratoryis that the putative cardiovascular healthbenets associated with plant food-rich dietscan be attributed in part to their avanolcontent. Dr. Keens group has over 600 peer-reviewed scientic papers in the above areas.

attacHMEnt a

BIoGrapHIcal SkEtcHES oF tHE panElIStS

15


16/20

d. Js rihs

Jason Richardson, M.S., Ph.D. is an AssistantProfessor in the Department of Environmentaland Occupational Medicine at Robert Wood

Johnson Medical School and Resident Memberof the Environmental and Occupational HealthSciences Institute. He received his M.S. andPh.D. degrees from Mississippi State Universitywhere he conducted research on mixturesof organophosphate pesticides and thedevelopmental neurotoxicity of organophosphatesduring critical periods of development. He thencompleted postdoctoral training in MolecularNeuroscience and Neurotoxicology at EmoryUniversity. His research at EOHSI focuses onthe role of environmental exposures during

development and how such exposures interactwith genetic susceptibility to produceneurological disease.

d. r rihs

Dr. Richardson is the Dow Professor ofToxicology and Associate Professor of Neurologyat the University of Michigan School of PublicHealth. He received his B.S. (magna cum

laude) in Chemistry from Wichita StateUniversity. Upon achieving Ph.D. candidacy inChemistry at SUNY Stony Brook, he transferredto Harvard, where he earned the Sc.M. andSc.D. degrees in Physiology/Toxicology. Afterpostdoctoral work in Neurochemistry at theMedical Research Council Toxicology Unit inCarshalton, England, he joined the Universityof Michigan as Assistant Professor of Toxicology.Apart from sabbatical leaves at Warner-Lambert/Parke-Davis (now Pzer) in Ann Arbor and theUniversity of Padua in Italy, Dr. Richardson hasbeen based at Michigan, where he has risenthrough the ranks to full professor. During1993-1999 he served as director of theToxicology Program and in 1998 he wasappointed as the Dow Professor of Toxicology.He is board-certied by the American Board ofToxicology (DABT). His research has focusedon mechanisms of acute and delayed

neurotoxicity of organophosphorus compounds.Currently he uses kinetics, molecular modelingand mass spectrometry to understand interactionsof toxicants with target macromolecules and todevelop biomarkers of exposure, toxicity and

disease.

d. k rm

Dr. Rozman is a Professor of Pharmacology,Toxicology & Therapeutics at the KansasUniversity Medical Center. He holds a Ph.D.from the University of Innsbruck in Organicand Pharmaceutical Chemistry. He is aDiplomate of the American Board of Toxicology

and a member of many journal editorialboards. Dr. Rozmans research is aimed atelucidating the mechanism of toxicity ofchlorinated aromatic hydrocarbons (CAH) andrelated compounds. The cause of2,3,7,8-tetrachlorodibenzo-p-dioxin-induceddeath (and related compounds) in rats is acombination of appetite suppression andinhibition of gluconeogenesis, whereas in miceit appears to be inhibition of gluconeogenesisalone, leading to a lethal hypoglycemia.Currently three lines of research are beingpursued: 1) elucidation of the molecularmechanism(s) of action leading to CAH-inducedenzyme inhibition; 2) investigation of thesubchronic and chronic toxicities of TCDD andits higher chlorinated homologues as well asother heterocyclic analogues such as chlorinatedphenothiazines (CPT), and 3) studying femalereproductive toxicity of both CAH and CPT. Dr.Rozman has studied chlorinated pesticidesextensively such as DDT, hexachlorbenzene,pentachlorophenol, dieldrin, heptachlor,

chlordane and more. He has published morethan 30 original manuscripts on these topicsand has written many book chapters and reviewarticles on chlorinated pesticides as well as onorganophosphates.

16


17/20

Bg pesiie regi

EPA regulates pesticides under the FederalInsecticide, Fungicide, and Rodenticide Act(FIFRA) and the Federal Food, Drug, andCosmetic Act (FFDCA). These acts weresignicantly amended in 1996 by the FoodQuality Protection Act (FQPA). FQPA was, atleast partly, motivated by the National ResearchCouncils (NRCs) 1993 report Pesticides in theDiet of Infants and Children, whichrecommended changes to EPAs risk assessmentmethods for pesticide residues on food,

particularly to provide better protection forinfants and children. FQPA called for enhancedstringency in the system of regulation forpesticides and adopted a reasonable certaintyof no harm standard.

The EPA regulates the residues of pesticides onfood commodities using an extensive riskassessment process, with two key elements: (1)characterization of toxicity through an extensivebody of required tests, and (2) estimation ofdietary exposure through the use of models

coupling data on food consumption with dataon pesticide residues from eld trials, monitoringdata, etc.

tii tesig

The EPA requires more toxicity data foragricultural pesticides of conventional chemistrythan any for other type of chemical. The data

requirements for pesticides are detailed in 40CFR Part 158. The required toxicity tests forthese pesticides used on food include:

Acute oral toxicity rat

Acute dermal toxicity Acute inhalation toxicity rat

Primary eye irritation rabbit

Primary dermal irritation

Dermal sensitization

Acute neurotoxicity rat

90-day oral rodent

90-day oral non-rodent

21/28 day dermal

90-day neurotoxicity

Chronic oral rodent

Carcinogenicity two rodent species

Prenatal developmental toxicity

Reproduction and fertility effects

Bacterial reverse mutation assay

In vitro mammalian cell assay

In vivo cytogenetics

Metabolism and pharmacokinetics

Immunotoxicity

All of these studies are conducted under GoodLaboratory Practices (GLP) and the data arereviewed by EPA before they are judgedacceptable for risk assessment.

As listed in 40 CFR 158.500, there several othertoxicity tests that EPA can conditionally requireif needed to rene the risk assessment (e.g.,

developmental neurotoxicity). Also, manyregistrants voluntarily conduct additionaltoxicity studies to rene the risk assessment oftheir chemicals or to fulll requirements inother countries. Also, at its discretion, EPA canuse open literature data to rene theassessment.

attacHMEnt B

EpaS rEGulatory procESS For pEStIcIdE rESIduES on Food

17


18/20

deeme tii

reeee ves

Following the receipt, review and acceptanceof the toxicity data by EPA, toxicity reference

values are derived for acute and chronicexposure durations, and for lifetime cancer risk,if the pesticide is found to be a carcinogen. EPAuses standard methods to calculate the toxicityreference values, except that they must add anadditional 10-fold safety factor to protectchildren, unless the available data show thatsome other factor is more appropriate (seebelow).

The rst step in the process is the determination

of the point-of-departure for risk assessment.Historically, the point-of-departure was a noobserved adverse effect level (NOAEL) from atoxicity study of appropriate duration. However,EPA is moving away from the use of NOAELswhen possible, and, instead, derivingbenchmark doses (BMDs). As an example, EPAhas derived BMDs for cholinesterase-inhibitors,including organophosphates and N-methylcarbamates. To estimate the BMD for this class,EPA rst nds whether the brain or red bloodcell (RBC) compartments are more sensitive.

The BMD for the point-of-departure is usuallychosen as the estimated dose that causes a 10%inhibition of either brain or RBC cholinesterase,whichever gives a lower result. This approach ismore conservative than other agencies such asthe World Health Organization (WHO) whichrecommends a 20% inhibition for the point-of-departure.

EPA applies various uncertainty factors to thepoint-of-departure, generally including a

default 10-fold factor for animal-to-humanextrapolation (interspecies variation) and adefault 10-fold factor for intraspecies variation.Chemical-specic data, when available, wouldprompt the application of chemical-specicuncertainty factors. EPA may also applyadditional factors for database deciencies orfor extrapolation from subchronic to chronicexposures.

One of the most signicant changes mandatedin FQPA was the obligation of the agency toapply an additional default safety factor of 10for the added protection of infants and children.The FQPA 10X factor can be adjusted if on

the basis of reliable data, such margin will besafe for infants and children. As an example, ifit can be shown that there is no difference intoxicity for infants and fetuses, compared toadults and there are no databases deciencies,then the FQPA factor may be reduced to as littleas 1X.

For cancer risk assessment, as a default, EPA

typically uses a linear, no-threshold dose-response model to estimate a unit risk (orpotency) factor, based on tumor rates in theanimal studies. The unit risk factor can bemultiplied by a lifetime average exposure toestimate a lifetime risk. In many cases,depending upon the number and nature of theobserved tumor types and number of speciesshowing a positive carcinogenic response, amargin-of-exposure approach to the quantitativerisk assessment may be preferred.

Ese assessme

EPA estimates dietary exposure to pesticidefood residues using residue data collected ineld trials, post-harvest, or in market basketsurveys, in combination with data on foodconsumption.

Pesticide registrants obtain registrations for apesticide on a crop-specic basis. Therefore,for each crop that a pesticide is used on, theregistrant must submit eld trial data thatinclude measurements of pesticide residues onthe commodity following an application at themaximum application rate and minimum

18


19/20

pre-harvest interval that will be allowed. Fromthe eld trial data, a tolerance is established foreach crop-pesticide combination. The tolerancerepresents an upper-bound estimate of thepesticide residue concentration on the crop;

the toxicity of the pesticide does not play a rolein establishing a tolerance, although it is afactor in whether or not the tolerance isapproved. If a food is found with residuesexceeding the tolerance or with residues of anunapproved chemical, the food is consideredadulterated.

EPA uses residue data from the PDP and FDAsmonitoring programs in its higher-tier riskassessments.

EPA characterizes food consumption usingUSDAs Continuing Survey of Food Intake byIndividuals (CSFII). The CSFII is a survey of thefood intake of more than 20,000 individuals.These data are used to provide a distribution ofthe intake of a large range of different fooditems, and the data are divided into differentage and gender categories to derive separatepesticide residue exposures across thesecategories. Therefore, food intake specic toinfants and children are used, as well as womenof child-bearing age (a separate risk assessmentmay be conducted for this subset of adultfemales if there are reproductive ordevelopmental effects). The software programsused by EPA to perform these calculationscontain recipes for processed foods whichallow users to estimate the pesticide residues inprocessed food based on the differentingredients. Also, EPA has published defaultprocessing factors which provide estimates ofthe effect of processing on pesticide residues

compared to raw commodities. For higher tiersof risk assessments, registrants may also conductadditional chemical-crop specic processingstudies.

There are four tiers of dietary risk assessment,each with increasingly complexity:

1. Tier 1: Tolerance levels are used to estimateresidue levels; 100% of the crop is assumedto be treated with the pesticide (usually avery conservative assumption); defaultprocessing factors.

2. Tier 2: Either the tolerance or the highestresidue level from the eld trial is used (forcertain foods, average eld trial residuesmay be used); 100% crop treated; chemical-crop specic processing factors are used.

3. Tier 3: The entire distribution of eld trialdata and/or the PDP survey data can beused and adjusted for the percent croptreated (i.e., the percent of a given croptreated with a pesticide.

4. Tier 4: A market basket survey is conductedand used for risk assessment.

The tiering system provides a variety ofapproaches from easily applied methods to(lower tiers) that give conservative results (i.e.,tend to overestimate risk) to more sophisticatedmethods that require more effort but moreclosely approximate reality.

In all cases, software is used to estimate thepesticide exposure for each individual or foreach population group in the CSFII based ontheir individual consumption data or on averageconsumption estimates for the population. Theend result is a distribution of exposures fordifferent age groups.

19


20/20

ris assessme

EPA conducts a risk assessment by comparingthe Population Adjusted Dose (PAD) with thepesticide residue dose estimates. The PAD is

the reference dose (point of departure dividedby uncertainty factors) divided by the FQPAfactor. The PAD and RfD are the same if theFQPA factor is reduced to 1X. For an acute riskassessment, EPA takes the 99.9th percentile ofthe distribution across the individuals in theCSFII for each subpopulation and comparesthat with the acute PAD6.

For chronic risk assessment, EPA compares themean pesticide residue dose estimate with the

chronic PAD. For cancer risk assessment (ifapplicable), EPA estimates a lifetime cancer riskby multiplying the unit risk estimate by the doseestimate or calculates the size of the margin-of-exposure.

If, after all renements in the risk assessmentare complete, there is an exceedance of eitherthe acute or chronic PAD, or if the cancer riskis greater than 1 in a million or exceeds anacceptable margin-of-exposure, adjustmentsmust be made. The adjustments may include a

deletion of one or more crops from the label(thus disallowing use on those crops) or achange in the use pattern(s), including theapplication rate, number of applications,interval between applications, or pre-harvestinterval. After the adjustments are made, theassessment is rerun to determine if thereasonable certainty of no harm standard hasbeen met. This process is repeated until thestandard is met, the use(s) is/are cancelled, orthe registration for the chemical is discontinuedby the registrant or cancelled by EPA.

6 Since the Tier 1 assessment is so conservative, a 95th percentile is used.

Documents

56Expert Panel Report Final With Graphics