Worst First:
High-Risk Insecticide Uses, Children's Foods and Safer Alternatives

Prepared by Consumers Union Washington, D.C.
September, 1998
This report is also available in pdf format
Return to Index

The Risk Cup Overfloweth

The FQPA commands the EPA to reduce risks from pesticides in the diet to levels that have a "reasonable certainty of no harm" to public health, and to ensure that childrenís health is protected.

How much reduction in exposure will be required to meet that goal? EPA canít say yet. Especially for chemicals with a common toxic mechanism like the OP and carbamate insecticides, EPA must first determine what a safe overall exposure to members of the class is. Then the agency must determine how much current risk exists from cumulative exposure to all members of the category through dietary residues and non-food exposures. The degree to which current cumulative risk exceeds the safe exposure is the amount of risk reduction needed.

EPA has coined the term "the risk cup" to describe the acceptable risk level; itís the sum of exposures that, together, donít exceed a maximum safe daily intake for kids. Imagine a container, or cup, with a fixed capacity, and think of each individual pesticide use as creating a risk of some size that fills part of the cup. The risk cup for OP insecticides, for instance, may allow a good number of crop-specific uses, as long as aggregate exposure and risk from all those uses (and from other permitted non-dietary exposure to OPs) does not exceed the maximum safe level, i.e., doesnít make the risk cup overflow.

Thereís a lot of uncertainty so far over just how big the risk cup (or cups) for the OPs and carbamates will be. But based on the work done by the NRCís Committee on Pesticides in the Diets of Infants and Children (NRC 1993), and earlier this year by the Environmental Working Group (EWG 1998), dietary exposure currently appears to exceed a safe level by a wide margin. In a recent presentation to its Tolerance Reassessment Advisory Committee, EPA reported the results of preliminary analyses suggesting that about 20 OP insecticides exceed safe exposure levels on their ownówithout considering their common toxic mechanism.

Does that mean that all OP and carbamate uses must be eliminated, or that every tolerance for use of these insecticides on foods has to be revoked or drastically reduced? No, not necessarily. Some uses pose far greater risks than others. For example, insecticides applied early in the growing season often leave no detectable residues on the harvested crop, and can reasonably be assumed to pose significantly smaller risks than applications later in the season. Other uses, on crops that kids seldom eat, also contribute less to risk, at least for infants and children. On the other hand, applications close to harvest time on foods that are prominent in childrenís diets are most likely to contribute heavily to overall dietary exposure and risk.

EPAís challenge is to manage the aggregate risk by eliminating pesticide uses that create the biggest risks, making room in the risk cup for other lower-risk uses of economically valuable chemicals. The FQPA in fact requires EPA to prioritize among risks and to regulate the "worst first."

High-Risk Insecticide Uses

EPA has already made it clear that, collectively, the OPs and carbamates fall into the high-risk category. But different uses pose different risks. How can big risks be sorted out from little risks in this category, to determine what uses fit within the risk cup?

Our analysis identifies high-risk OP and carbamate uses. We have focused on the central mandate of FQPAóprotecting childrenís healthóand defined high-risk based on three factors: The role of specific foods in childrenís diets; the occurrence of residues of specific OP and carbamate insecticides on or in foods kids eat a lot of; and the relative toxicity of the residues. Insecticide uses we consider "high-risk" are those that frequently leave residues of comparatively toxic members of these chemical families in foods kids consume a lot of.

We have identified 40 specific crop-chemical uses that are "high risk" by our criteria; theyíre listed in Table 2.1. These 40 uses are a small fraction of

_____________________________________________________________

Table 2.1. The Worst 40: High-Risk Insecticide-Crop Uses
___________________________________________________________

Fruits Vegetables

Crops/Foods Insecticides Crops/Foods Insecticides

Apples Azinphos-methyl Green Beans Methyl Parathion

Chlorpyrifos Methamidophos

Methyl Parathion Dimethoate

Dimethoate Acephate

Carbaryl Carbaryl

Oxamyl

Pears Azinphos-methyl Peas Dimethoate

Methyl Parathion Acephate

Phosmet

Carbaryl

Oxamyl

Peaches Azinphos-methyl Potatoes Methamidophos

Chlorpyrifos Aldicarb

Diazinon

Methyl Parathion

Phosmet

Formetanate HCL

Aldicarb

Carbaryl

Grapes Azinphos-methyl Tomatoes Azinphos-methyl

Chlorpyrifos Chlorpyrifos

Formetanate HCL Methamidophos

Dimethoate

Methomyl

Carbaryl

Oranges Methidathion

Chlorpyrifos

Carbaryl

_____________________________________________________________________

the estimated 300 current food-production uses of OPs and carbamates that entail applications to more than 1 percent of national crop acreage (and an even smaller fraction of the estimated 700 registered, or legally permitted, uses). But, collectively, they account for a disproportionately large share of total risk associated with insecticide use on food crops. These "Worst 40" uses are clear top priorities for EPAís risk management efforts under FQPA.

As Table 2.1 shows, the 40 specific uses involve 14 insecticides used in various combinations on nine food crops. The rest of this chapter explains how we identified them as the "Worst 40."

Identifying Foods Consumed in Disproportionate
Amounts By Infants and Children 

As stated in Pesticides in the Diets of Infants and Children (NRC 1993), "ChildrenÖconsume more calories of food per unit of body weight than do adults. At the same time, infants and children consume far fewer types of food than do adults." Differences in dietary exposure attributable to these two combined factors are among the most important reasons children face greater health risks from pesticides than adults do.

We began our screening for high-risk insecticide uses by looking at food intake data, to determine which foods kids eat in significant quantities. The 1993 NRC report identifies 23 foods or food groups that each make up more than 1 percent of the diets of children at some point from infancy through age 12, based on a national food consumption survey conducted in 1977-1978. (Though dated, these survey data are the most comprehensive available.) Table 2.2 lists these 23 foods, and shows how much each contributes to the diet for children of different ages.

A more recent survey (USDA 1996) focused on food intake by children showed largely the same food consumption patterns, but noted some changing trends. The survey found that:

• Kids are consuming more beverages, grain-based snacks and combination foods like pizza, and eating away from home more.
• Kidsí milk and fat consumption is declining.
• Consumption of conveniently packaged drinks based on apple, grape and mixed fruit juices is rising markedly. Soft drinks account for the rest of the increase in beverage consumption.

Table 2.2. Foods and Crops That Account for More Than 1 Percent of the Diet of Infants and Children
Food		Nursing Infants		Non- nursing Infants	One to Six Year Olds	Age Seven to Twelve	Age 13 to 19 Years
		Percent of Average Diet by Age Group
Milk and Dairy		36%		58%	44%	42%	42%
Fruits
Apples		15.5%		7.6%	4.1%	2.1%	1.6%
Peaches		4.1%		2.3%	<1%	<1%	<1%
Pears		4.3%		1.9%	<1%	<1%	<1%
Orange Juice		4.5%		3.2%	5.7%	4.6%	4.2%
Bananas		2.7%		1.1%	<1%	<1%	<1%
Total Fruits		31.1%		16.1%	11.3%	8.2%	7.3%
Grain Based Products
Wheat Flour		<1%		1.0%	4.6%	5.8%	6.1%
Oats		1.2%		<1%	<1%	<1%	<1%
Milled Rice		1.8%		1.5%	<1%	<1%	<1%
Total Grains		3.5%		3.0%	5.60%	6.8%	7.1%
Vegetables
Tomatoes		<1%		<1%	1.4%	1.7%	2.1%
Carrots		3.3%		1.6%	<1%	<1%	<1%
Peas		1.3%		<1%	<1%	<1%	<1%
Beans		1.3%		<1%	<1%	<1%	<1%
Potato		<1%		<1%	3.9%	4.6%	5.3%
Sweet Corn		<1%		<1%	<1%	1.1%	1.1%
Total Vegetables		7.4%		4.1%	7.3%	8.9%	10.0%
Other
Beef (lean plus fat)		3.0%		1.5%	4.6%	5.7%	7.1%
Soybean Oil		1.1%		1.5%	1.0%	1.2%	1.4%
Coconut Oil		<1%		1.4%	<1%	<1%	<1%
Cane Sugar		<1%		<1%	3.1%	3.4%	3.7%
Eggs		<1%		<1%	1.4%	1.4%	1.6%
Beet Sugar		<1%		<1%	1.4%	1.5%	1.7%
Pork		<1%		<1%	1.2%	1.3%	1.8%
Chicken		<1%		<1%	1.4%	1.4%	1.6%
Total Other		7.1%		6.9%	14.6%	16.4%	19.4%
Note: All foods @ <1% of consumption assumed to account for 0.5% in estimating totals.
Source: Table 5-6, Pesticides in the Diets of Infants and Children (NRC 1993).

 A few points shown in Table 2.2 are worth highlighting. Food consumption patterns change dramatically as children pass through infancy and the first few years of life, and continue to change as they mature into adults.

Milk and dairy products are the single biggest dietary component for infants and children. Next to milk, Table 2.2 shows that orange juice makes up the greatest percentage of the diet for children ages one to six years. Apples, apple juice, peaches and pears are consumed by children under age one in amounts that are five to 15 times the national average intake per unit of body weight, and non-citrus juices now account for 6 percent of total daily food intake of three to five year-oldsóabout three times average intake (USDA 1996). Kids older than 5 years switch from apple to citrus as the dominant juices in their diets, and non-milk beverages displace milk as kids reach their teens. As children grow older, vegetable consumption increases, led by potato and tomato-based products, while fruit intake declines.

Our analysis of the foods most likely to contribute to dietary insecticide exposure in children narrowed the focus to nine high-intake foods. They include one not listed in Table 2.2ógrapesóand eight others that do appear in the table, all fruits and vegetables. The nine foods are: apples, pears, peaches, grapes and oranges; peas, green beans, potatoes and tomatoes.

In selecting these nine foods, we relied on two additional kinds of data: The frequency of detection of insecticide residues in the foods, and the relative toxicity of residues commonly detected in each food.

Grapes made our list, although they accounted for less than 1 percent of childrenís diets in the 1977-78 food survey, because their consumption in fresh, processed and juice forms has grown rapidly since then. An analysis by the Environmental Working Group published earlier this year showed that grape-based foods contribute significantly to excessive OP exposure in childrenís diets (EWG 1998).

Foods in Table 2.2 that are not among our chosen nine high-risk foods are typically less likely to contain insecticide residues than foods we selected. Carrots generally have very rare, low residues of OPs and carbamates; edible portions of sweet corn and bananas are also comparatively "clean," although more residues are found on inedible outer husks. OP and carbamate residues are almost completely absent from milk and dairy products, meats, vegetable oils, and sweeteners.

We also excluded all grain-based products, for somewhat different reasons. Most growing-season insecticide uses on these crops leave no detectable residues when foods are consumed. On the other hand, post-harvest insect control, during storage, transportation and processing of grains, often does leave residues, which contribute significantly to childrenís overall dietary exposure (EWG 1998). However, our focus in this report is on insecticide use in farm production and on safer alternative pest management choices that growers can use.

Identifying the Most Toxic Insecticides Found in Foods

Pesticides are not all equally toxic, and the degree of risk posed by dietary insecticide residues depends on the toxicity of the individual insecticides, as well as on the frequency of occurrence and level of residues present. The second step in our screening for high-risk insecticide uses was to look at the comparative toxicity of different OP and carbamate compounds commonly found in foods.

Pesticide toxicity is measured in many ways. At least two kinds of adverse effects are considered: acute toxicity, in which effects on one or more body systems are detected immediately following exposure; and chronic toxicity, in which effects occur only after longer-term, lower-level exposure, or long after an acute exposure. Typically, toxicity is tested in a variety of animal experiments, and data from human (usually, occupational) exposure are also relied on when available.

An index of how toxic an insecticide is, widely used by risk assessors and regulators, is the reference dose, or RfD for short. The RfD is an exposure level, expressed in milligrams of chemical per kilogram of body weight of the exposed individuals per day, estimated to pose no appreciable risk of adverse effect in people. Most chemicals have an RfD for chronic effects, and some may have one for acute effects, as well.

Toxicologists calculate an RfD by first determining the lowest level of exposure to a chemical that produced an adverse effect in a well-designed animal study. The next exposure level below thatóthe highest dose that produced no observable, statistically significant adverse effect in the group of animals exposed to itóis called the No Observable Adverse Effect Level, or NOAEL. The NOAEL is typically divided by a "safety factor"óranging from 100 to 1,000 depending on the extent and quality of available dataóto produce the RfD, the estimated safe daily dose for humans, including of course vulnerable population groups like children.

RfDs may be based on any of a wide array of toxic effects; ordinarily, the adverse effect that occurs at the lowest level of exposure is deemed most critical, and is the basis for the RfD. For most insecticides, especially OPs and carbamates that share a common mechanism of toxicity, effects on the nervous system are the uniform basis for RfDs. This allows straightforward comparisons of relative toxicity for members of these families.

Table 2.3 displays chronic RfDs for the OP and carbamate insecticides. These RfDs are based on the EPAís risk assessments, and are the agencyís current official estimates of the "safe" dose for each chemical listed.

Since an RfD is derived from the amount of a chemical required to produce an adverse effect, the smaller the RfD, the less of a substance needed to have toxic effects, and the more hazardous the chemical. Among the OPs shown in Table 2.3, there is a 2,000-fold difference in chronic toxicity between methyl parathion and malathion. Among carbamates, oxamyl and aldicarb are about 70 and 14 times as toxic, respectively, as carbaryl, based on comparative chronic RfDs for the three insecticides. Comparing across the two groups, methyl parathion is 700 times as toxic as carbaryl.

The most toxic insecticides are very toxic indeed. The cumulative toxicity of the OPs found in apples and in peaches is such that a child who eats three-fourths of an apple, or a whole peach, has roughly a one-in-four chance of exceeding the RfD (i.e., the safe daily intake) for OPs, just from eating that one food item (EWG 1998).

Some of the most toxic OP and carbamate insecticides are severely restricted by EPA and may not legally be used on most crops. They are rarely found in the foods they can be applied to. Other members of these two chemical families, such as acephate, azinphos-methyl or carbaryl, are much less toxic but are much more widely used, and consequently may contribute significantly to the overall risk from insecticide residues in foods eaten widely by children.

Such wide differences in toxicity within a chemical family make it clear that some insecticides of each type are far riskier, and some are far less risky, than others.

Table 2.3. Comparative Chronic Toxicity of the Organophosphate and

Carbamate Insecticides

OPP/EPA Chronic Reference Dose

 

Insecticide (mg/kg body weight) .

 

Organophosphates

Methyl Parathion 0.00002

Profenofos 0.00005

Terbufos 0.00005

Pirimiphos methyl 0.00008

Dicrotophos 0.0001

Ethoprop 0.0001

Fenamiphos 0.0001

Dichlorvos 0.00017

Chlorpyrifos 0.0003

Disulfoton 0.0003

Ethyl parathion 0.00033

Dimethoate 0.0005

Ethion 0.0005

Oxydemeton-methyl 0.0005

Phorate 0.0005

Chlorethoxyfos 0.0006

Diazinon 0.0007

Methamidophos 0.001

Acephate 0.0012

Azinphos-methyl 0.0015

Methidathion 0.0015

Fonofos 0.002

Naled 0.002

------------------------------------------------------------------------------------------------------------

Lower Risk OPs

 

Phosmet 0.003

Sulprofos 0.003

Chlorpyrifos-methyl 0.01

Malathion 0.04

 

 

Carbamates

Oxamyl 0.0002

Aldicarb 0.001

Formetanate HCL 0.002

Carbofuran 0.005

------------------------------------------------------------------------------------------------------------

Lower Risk Carbamates

 

Methomyl 0.008

Carbaryl 0.014

Thiodicarb 0.03

Fenoxycarb 0.08

 ___________________________________________________________

EPA will need to draw such distinctions in setting regulatory priorities. In Table 2.3, we have indicated where we think reasonable lines might be drawn between high-risk and lower-risk members of these two insecticide groups.

Highest-Risk Food/Insecticide Combinations

How important a particular foodís contribution is to childrenís overall risk from insecticide exposure depends on which insecticides are used on the crop, and on the extent to which residues remain in the food as eaten. As we explained earlier, we relied on qualitative comparisons of residue prevalence to narrow the list of high-intake foods down to nine, and we collected data on the relative toxicity of all the OP and carbamate insecticides frequently found on foods. Then we took a closer look at residues in the nine foods, to affirm that they belong in the "high-risk" category and to identify the specific insecticide uses that seem most likely to drive childrenís dietary exposure and risk.

The data we examined come from two major Federal Government pesticide residue testing programs. The U.S. Department of Agriculture (USDA) tests foods for pesticide residues, with emphasis on monitoring foods eaten in quantity by children. The USDA Pesticide Data Program (PDP) tests foods "as eaten"ówashed, peeled, and for processed foods, cooked.

But the PDP tests only some 14 high-consumption foods. To get a broader picture of residues in foods, we supplemented our analysis of PDP data by examining results from the U.S. Food and Drug Administrationís Pesticide Surveillance and Monitoring Program. The FDA carries out comprehensive annual testing to enforce tolerances for residues, collecting samples just after the foods leave the farm. The samples are tested "as is," without washing, peeling or cooking. These methods find residues somewhat more often, and at higher levels, than the PDP testing finds. Still, the FDA data are among the best available to assess residue levels in foods not tested by the PDP.

Table 2.4 presents an overview of OP and carbamate residues in our nine high-consumption childrenís foods. The table shows the percent of samples of each food item tested by USDA or FDA that were positive for each listed insecticide. Eleven OPs and five carbamates were detected frequently in the nine foods, and are listed. Other members of these insecticide families were found so infrequently and at such low levels that they contribute modestly at most to dietary exposure, and we have excluded them from this analysis.

Table 2.4. Frequency of Detection of Organophosphate and Carbamate Insecticide Residues in High-Consumption Childrenís Foods

_____________________________________________________________________

Vegetables Green Beans Peas Potatoes Tomatoes

Carbamates

Aldicarb 0.2 19.0

Carbaryl 11.9 2.0 1.1

Formetanate HCl

Methomyl 0.9

Oxamyl 1.1

_____________________________________________________________________

Data for apples and pears are split into fresh and juice food forms because the residue data are reported that way, but we consider such paired subsets to be single food-insecticide combinations, since both represent the same crop. A blank cell in the table means the insecticide in that row was not detected in the food at the top of the column. Seventy food-insecticide combinations fall into this non-detected category; residues were found in 74 cases.

To determine which of those 74 positive combinations are "high-risk" food- insecticide uses, we applied two criteria: frequency of residue detection, and relative toxicity of the individual insecticides. The "Worst 40" uses are in bold type in Table 2.4. Here are the criteria we used to select them:

• For most OP and carbamate insecticides, those uses that leave residues in 2 percent or more of PDP samples, or in 4 percent or more of FDA surveillance samples, are classed as "high-risk" combinations.
  
• For the "lower-risk" OPs phosmet and malathion, and for the carbamates carbaryl and methomyl, crop uses that leave residues in more than 5 percent of the tested samples are classed as "high-risk" combinations.

Among the Worst 40 insecticide-food combinations (excluding the juices), 21 have detection frequencies of 10 percent or higher and 10 combinations exceed 25 percent frequency. In the worst case, children who eat apples are likely to be exposed to azinphos-methyl 54.5 percent of the time.

As Tables 2.1 and 2.4 show, "high-risk" uses are concentrated in five foods: peaches (8), apples (6), grapes (6), green beans (5) and pears (5) account for 30 of the Worst 40 insecticide-food combinations. The most worrisome OPs are azinphos-methyl and chlorpyrifos, each in five of the nine foods, and methyl parathion, in four. Among carbamates, aldicarb is the most serious concern because of its high acute toxicity, its presence in potatoes, a major food, and recent increases in the percent of acres treated. Carbaryl is found frequently in six of the nine foods, but is far less toxic than aldicarb.

For a more detailed and elegant analysis of pesticide residues in childrenís diets, and one that also supports our selection of high-risk insecticide-food combinations, we refer readers to the Environmental Working Groupís 1998 report, "Overexposed: Organophosphate Insecticides in Childrenís Food" (EWG 1998). We also invite readers to visit our project web site for our own more detailed analysis (http://www.ecologic-ipm.com).