WRITTEN OBJECTIONS TO VINCLOZOLIN TOLERANCE FOR SNAP BEANS
and
Al Meyerhoff
Natural Resouces Defense Council
71 Stevenson Street, Suite 1825
San Francisco, CA 94105
(415) 777-0220
On Behalf Of
Natural Resources Defense Council
American Federation of Labor and Congress of Industrial Organizations
Northwest Coalition for Alternatives to Pesticides
Pineros y Campesinos Unidos del Noroeste
September 15, 1997
WRITTEN OBJECTIONS TO VINCLOZOLIN TOLERANCE FOR SNAP BEANS
On behalf of Natural Resources Defense Council ("NRDC"), American Federation of Labor and Congress of Industrial Organizations ("AFL-CIO"), Northwest Coalition for Alternatives to Pesticides ("NCAP"), and Pineros y Campesinos Unidos del Noroeste ("PCUN"), we hereby submit the following written objections in accordance with 40 C.F.R. §§ 178.20-.37 to the tolerance granted for residues of vinclozolin on snap beans. The submitters will be referred to as NRDC throughout these objections.
SUMMARY
NRDC objects to the following provisions of the tolerance regulation:
Pursuant to 40 C.F.R. § 178.27, the submitters also seek an evidentiary hearing on the following issues:
RELIEF
In terms of the relief sought, NRDC asks EPA to rescind the tolerance granted for vinclozolin residues on snap beans. EPA should also take immediate action to revoke other tolerances for vinclozolin, iprodione, and procymidone on foods to the extent necessary to bring tolerances for those substances in line with FQPA standards.
FEE WAIVER
Under 40 C.F.R. § 180.33(i), objections to a tolerance must be accompanied by a filing fee of $3,025. The Administrator may waive or refund such fees when such a waiver or refund will promote the public interest. Id. § 180.33(m). While ordinarily a $1500 fee must accompany any request for a fee waiver or refund, this fee is not imposed on any person who has no financial interest in the matter at issue. Id.
NRDC is lodging the $3025 filing fee, but asks the Administrator to refund this amount in full. None of the submitters has any financial interest in this tolerance or this chemical. Instead, they seek to protect the public from the health risks presented by this pesticide. Imposing such a significant fee will deter interested members of the public from pursuing the objection process. Accordingly, waiving the
objection fee in this instance will promote the public interest in protecting public health.
Indeed, we urge the Administrator to amend EPA’s regulations to provide a blanket waiver in these instances to ensure that the current mandate to pay the $3025 and then seek a refund does not create insurmountable obstacles to future tolerance objections. Under the Freedom of Information Act, 5 U.S.C. § 552(a)(4)(A), fee waivers may be sought without any requirement that the fees be posted. This process ensures that the requirement to pay fees does not deter public interest requesters from using the Act. So too, EPA should ensure that its objection process is available to individuals and entities that have no financial stake, but who seek to promote the overall safety of our food supply.
BACKGROUND
I. THE RISKS PRESENTED BY VINCLOZOLIN
Vinclozolin is a fungicide produced under the trade name Ronilan by BASF Corporation. EPA has established tolerances permitting residues of vinclozolin on Belgian endive, cucumbers, dried prunes, grapes, grape pumace, kiwifruit, lettuce, onions, bell peppers, plums, prunes, raisins, raspberries, stonefruits, strawberries, and tomatoes. 40 C.F.R. § 180.380. Vinclozolin is also used on turf, which, at the time the tolerance petition was submitted, included residential lawns, parks, schools, and playgrounds.
Vinclozolin has been associated with numerous adverse health effects. Most serious, vinclozolin has been shown to disrupt the endocrine system, which regulates the release of hormones and other substances that control various functions, such as sexual development.
Scientific studies have demonstrated anti-androgen or demasculinizing effects from in utero and early post-natal exposures to vinclozolin. In one study in which vinclozolin was fed to rats during critical periods of fetal development, male offspring exhibited reduced anogenital distance, nipple development, and cleft phallus. Other forms of feminized genitalia were observed, including such malformations as vaginal pouches, undescended testicles, and reduced or absent prostate glands. Gray, Ostby & Kelce, "Developmental Effects of an Environmental Antiandrogen: The Fungicide Vinclozolin Alters Sex Differentiation of the Male Rat," 129 Toxicology & Applied Pharmacology 46-52 (1994).
Several peer-reviewed studies have linked vinclozolin’s anti-androgen effects to one of the metabolites produced as vinclozolin degrades. Kelce, Monosson, & Gray, "An Environmental Antiandrogen," 50 Recent Progress in Hormone Research 449 (1995); Wong, Kelce, Sar & Wilson, "Androgen Receptor Antagonist versus Agonist Activities of the Fungicide Vinclozolin Relative to Hydroxyflutamide," 270 Journal of Biological Chemistry 19998 (1995); Kelce, Monosson, Gamcsik, Laws & Gray, "Environmental Hormone Disruptors: Evidence that Vinclozolin Developmental Toxicity is Mediated by Antiandrogenic Metabolites," 126 Toxicology & Applied Pharmacology 276 (1994). Based on these and other studies, EPA has concluded that vinclozolin causes anti-androgen effects. See EPA Memorandum on Developmental & Reproductive Toxicity Peer Review of Vinclozolin (Nov. 8, 1994).
In addition, based on scientific studies, EPA had classified vinclozolin as a probable human carcinogen prior to this tolerance proceeding. EPA Office of Pesticide Programs List of Chemicals Evaluated for Carcinogenic Potential (Feb. 19, 1997).
II. VINCLOZOLIN’S REGULATORY HISTORY
Beginning in the early 1980s, the Oregon Department of Agriculture sought and obtained emergency exemptions each year for fourteen years for use of vinclozolin on snap beans. See Oregon Dep’t of Agriculture Request for Specific Exemption for Use of Vinclozolin Fungicide on Snap Beans, at 1-2 (Jan. 8, 1997) ("1997 Emergency Exemption Request"). During the 1980s, BASF Corporation had filed an application for registration of vinclozolin for use on snap beans, but that application was incomplete and did not permit EPA to make a registration determination. See 1997 Emergency Exemption Request at 2.
Since the late 1970s, Oregon growers have become dependent on vinclozolin to control gray and white molds on snap beans. The growers had shifted away from pole snap beans, which were harvested with manual labor, to bush snap beans that could be mechanically harvested, but the mechanical harvesting failed to distinguish between beans with and without mold. 1997 Emergency Exemption Request at 1-2.
When the molds become resistant to common fungicide combinations in the late 1970s, the growers began using vinclozolin under emergency exemptions. Prior to the enactment of the FQPA in August 1996, emergency exemptions could be issued for food uses for which there is no tolerance under the FFDCA. In other words, EPA could grant emergency exemptions for use of pesticides without determining whether it would be safe for consumers to eat the foods.
When the Oregon Department of Agriculture sought its fifteenth consecutive emergency exemption for the 1997 season, PCUN and NCAP filed comments with EPA objecting to this recurring use of the emergency exemption process for what had become a routine pesticide use. Presumably as a result of these comments, EPA and BASF Corporation abandoned the emergency exemption process and instead pursued a Section 3 registration.
This year, under either regulatory procedure, EPA could not permit continued use of vinclozolin on snap beans without first issuing a FFDCA tolerance for residues of the pesticide on the beans. Under a longstanding regulation, EPA cannot issue a FIFRA registration for a food use unless all required food tolerances are in place. 40 C.F.R. § 152.112(g). Under FQPA, EPA must establish tolerances that satisfy FQPA’s safety standard for pesticide chemical residues resulting from pesticide use under emergency exemptions. 21 U.S.C. § 346a(l)(6); see also Letter to State Secretaries of Agriculture from EPA Office of Prevention, Pesticides & Toxic Substances (Oct. 24, 1996) (notifying states that tolerances are required for emergency exemptions).
On May 30, 1997, EPA issued a registration for use of vinclozolin on snap beans. However, in clear violation of its own regulation, EPA did so without first ensuring that the required food tolerance was in place. In fact, EPA refrained from issuing the food tolerance because certain required determinations had not yet been made.
On July 10, 1997, PCUN and NCAP filed a lawsuit challenging EPA’s approval of the vinclozolin registration without the required food tolerances being in place. PCUN had learned of the registration not from any regulatory process, but from observations by its representative of vinclozolin use in the fields, some of which are directly adjacent to migrant camps, thereby exposing children and families to the pesticide. Moreover, irrigators had not been informed of the vinclozolin use and, therefore, were unable to protect themselves from exposures by observing the re-entry periods or protective clothing requirements specified on the label. In the lawsuit, PCUN and NCAP sought a preliminary injunction directing EPA to withdraw the registration until the required tolerances had been issued.
Four days after filing of the lawsuit, EPA issued a tolerance for vinclozolin on snap beans, and that tolerance was published in the Federal Register on July 18, 1997. 62 Fed. Reg. 38,464. While the issuance of the tolerance ended the procedural violation at the heart of that litigation, the tolerance is the subject of these objections.
III. THE VINCLOZOLIN TOLERANCE-SETTING PROCESS
In March 1997, EPA published in the Federal Register BASF Corporation’s summary of its food tolerance application for vinclozolin on snap beans. 62 Fed. Reg. 13,000 (Mar. 19, 1997). EPA issued no proposal of its own, nor did it evaluate or offer its position on the points made in BASF Corporation’s summary.
BASF Corporation’s summary short-circuited some of the core determinations that must be made under the FQPA. PCUN, NCAP, NRDC and EWG submitted comments on BASF Corporation’s tolerance petition objecting to the petition as grossly inadequate under the FQPA. Letter to EPA (April 17, 1997). These comments highlight the petition’s many deficiencies:
(1) BASF Corporation failed to provide an adequate margin of safety for the particular risks presented to children;
(2) BASF Corporation urged EPA to disregard vinclozolin’s common mechanism of toxicity with two other fungicides and its carcinogenic effects;
(3) BASF Corporation failed to provide EPA any exposure information needed to assess the cumulative effects of exposures to two other fungicides that produce the same harmful metabolite that is produced by vinclozolin; and
(4) BASF Corporation undercounted existing exposures by failing to account for potential residues in drinking water and exposures through non-food uses of vinclozolin.
Not only did BASF Corporation’s summary truncate the statutorily mandated analysis, but it also relied on agency documents that had not been made available to the public. To obtain access to these documents, PCUN and NCAP filed a request under the Freedom of Information Act ("FOIA"), 5 U.S.C. § 552, seeking the petition, records cited in the petition, and other information in EPA’s possession needed to apply FQPA’s food safety standards. FOIA Request (April 17, 1997). As of this date, PCUN and NCAP have received no records in response to their FOIA request.
IV. THE TOLERANCE DECISION
In July 1997, EPA established a tolerance for residues of vinclozolin and its metabolite on snap beans at 2 parts per million (ppm). 62 Fed. Reg. 38,464 (July 18, 1997). The tolerance expires October 1, 1999.
A. EPA Failed To Use The Statutorily Mandated Tenfold Safety Factor To Protect Infants And Children.
The tolerance decision acknowledges that, under the FFDCA, as amended by the FQPA, "EPA shall apply an additional tenfold margin of safety for infants and children in the case of threshold effects to account for pre- and post-natal toxicity and the completeness of the database unless EPA determines that a different margin of safety will be safe for infants and children." 62 Fed. Reg. at 38,471. EPA further notes that it may use the standard 100-fold safety factor without an additional tenfold margin of safety "when EPA has a complete data base under existing guidelines and when the severity of the effect in infants and children or the potency or unusual toxic properties of a compound do not raise concerns regarding the adequacy of the standard margin/factor." Id. (emphasis added).
Ignoring the toxic effects and exposure prongs of this test, the tolerance decision states: "Given the completeness and reliability of the data base for vinclozolin, the Agency concluded that a margin of safety of 100 (without the additional safety factors suggested by the commentor) would be safe for children." 62 Fed. Reg. 38,466; id. at 38,471 ("Based on current toxicological data requirements, the database relative to pre- and post-natal toxicity is complete"). After reciting vinclozolin’s demonstrated anti-androgen effects, the strong basis for assuming such effects would occur in humans, and post-natal studies showing more severe effects in offspring than in parents, id. at 38,471, EPA concludes that "[t]he data on anti-androgen effects of vinclozolin are of sufficient quantity and quality that an additional uncertainty factor is not necessary in order to assure that infants and children will be safe from such effects." Id. at 38,466.
B. The Tolerance Decision’s Assumptions on Exposures are Not Conservative.
In terms of exposure, the tolerance decision is not conservative or complete in at least three respects.
First, it appears that EPA looked principally at adult consumption patterns, and focused almost exclusively on women of childbearing age, without adequately considering the particular consumption patterns of nursing infants, non-nursing infants, and children of various age groups. 62 Fed. Reg. at 38,466. This is despite the fact that severe post-natal effects have been observed in scientific studies.
Second, EPA abandoned a conservative assumption that it traditionally uses in estimating dietary exposure to a pesticide: that 100% of a crop that could be treated with a pesticide will, in fact, be treated. NAS, Regulating Pesticides in Food: The Delaney Paradox, at 59 (1987). In the past, EPA has claimed that this assumption leads to a worst-case exposure estimate and injects additional safety factors into its analysis.
In its vinclozolin tolerance decision, EPA refined its worst-case scenario by estimating the percentages of certain crops treated with vinclozolin, including some foods, such as cherries, raspberries, strawberries, and stone fruit, that have some of the highest tolerance levels. 62 Fed. Reg. at 38,469. EPA recognizes that it must ensure that exposure estimates based on actual percentages of crops treated do not under-estimate the exposures for any significant subpopulation, yet it purports to identify a high end exposure only for women of childbearing age, ignoring infants and children. Id.
Third, EPA ignored exposures from most non-food uses of vinclozolin. It did so because BASF Corporation has indicated that it plans to drop several food and non-food uses from its registration and label. Specifically, BASF Corporation proposes to delete from its vinclozolin registration all residential uses and to limit turf uses to commercial and industrial sites, golf courses, and greenhouses and nurseries, thereby eliminating turf use in parks, school grounds, and recreational areas. However, given that EPA has identified women of childbearing age as the population of greatest concern, it should not have excluded non-food exposures from turf simply because vinclozolin use on school lawns, playgrounds, and residential lawns will diminish over time.
C. EPA Failed To Consider the Full Cumulative Effects of Aggregate Exposures to Vinclozolin, Iprodione, and Procymidone.
FQPA requires EPA to consider "available information concerning the cumulative effects of [the pesticide] residues and other substances that have a common mechanism of toxicity." 21 U.S.C. 346a(b)(2)(D)(v). Despite the existence of two fungicides that produce the same metabolite that has been demonstrated to cause vinclozolin’s antiandrogen effects, EPA failed to incorporate exposures to these fungicides in assessing non-cancer risks.
EPA acknowledges that it has available data that enable it to resolve issues of common mechanisms of toxicity for some pesticide chemicals. It provides the following examples: "These pesticides include pesticides that are toxicologically dissimilar to existing chemical substances (in which case the Agency can conclude that it is unlikely that a pesticide shares a common mechanism of activity with other substances) and pesticides that produce a common toxic metabolite (in which case common mechanism of activity will be assumed)." 62 Fed. Reg. at 38,470.
EPA further recognizes that vinclozolin, iprodione, and procymidone share similar toxicological and structural properties, as well as a common metabolite, indeed the metabolite that has been associated with vinclozolin’s alarming anti-androgen effects in laboratory studies. Id; see supra at 4-5.
Under its own logic, EPA should then have assumed a common mechanism of toxicity. However, EPA asserts that it has no statutory obligation to conduct a cumulative effects analysis for the three pesticides in combination because it has not yet established policies on how to do so: "The Agency believes that ‘available information’ in this context might include not only toxicity, chemistry, and exposure data, but also scientific policies and methodologies for understanding common mechanisms of toxicity and conducting cumulative risk assessments." 62 Fed. Reg. at 38,470. In its view, the lack of official policies and methodologies justifies ignoring the available data: "The Agency has therefore concluded that it does not have sufficient available and reliable information concerning common mechanism of toxicity of vinclozolin, iprodione, and procymidone to analyze the common mechanism issue . . .." Id. at 38,466.
How EPA then proceeded is extremely puzzling. Its May 30, 1997 draft tolerance decision considered no cumulative exposures to iprodione or procymidone, and the final tolerance decision claims that "the supporting risk assessment was performed assuming that no common mechanism of toxicity exists." 62 Fed. Reg. at 38,466. In addition, nothing in the draft or final tolerance decision documents inclusion of exposures to iprodione and procymidone from food or water in EPA’s assessment of developmental, endocrine disruption, and other non-cancer risks.
However, as discussed above, EPA refrained from approving the tolerance in May 1997. The most significant difference between the draft and final tolerance decision is the addition of a discussion of cumulative exposures to all three pesticides as part of the cancer risk assessment, but still without any cumulative effects analysis for non-cancer risks. The final tolerance decision states that EPA is required to estimate risk for consumption of food and water containing the vinclozolin metabolite across vinclozolin, iprodione, and procymidone. 62 Fed. Reg. at 38,470. Nonetheless, EPA considered cumulative exposures to the common metabolite from all three pesticides only in its cancer risk assessment, and not in its estimates of other exposures and risks, even though it is the common metabolite that has been associated with vinclozolin’s anti-androgen effects.
The cancer risk assessment in the final tolerance decision uses a potency factor for a chemical that is structurally related to the vinclozolin metabolite, and it considers exposures to the metabolite from dietary and drinking water exposures to all three pesticides. Id. at 38,470-71. In that assessment, the total carcinogenic risk is notably estimated as 2.6 x 10(-6), which exceeds EPA’s negligible risk level of 1 in 1 million (1 x 10(-6)). See H. Rep. No. 104-669, pt. 2, 104th Cong., 2d Sess. at 41 (1996). EPA attaches little significance to this increased risk level because of "the uncertainties in both the toxicology used to derive the cancer potency of a substance and in the data used to measure and calculate exposure." 62 Fed. Reg. at 38,471. This is another aspect of the tolerance decision that errs on the
side of providing less, rather than greater, public health protection in the face of scientific uncertainties.
OBJECTIONS
These objections challenge EPA’s application of two key features of the new Food Quality Protection Act of 1996. Specifically, EPA failed: (1) to use the statutorily mandated tenfold safety factor to account for infants’ and childrens’ exposures to and toxic risks from vinclozolin; and (2) to incorporate into its assessment of noncancer risks the available information on cumulative exposures to other similar chemicals. Before describing each of these objections in further detail, the general regulatory framework is described below.
I. STATUTORY FRAMEWORK FOR REGULATING PESTICIDES IN FOOD
Pesticide use is regulated under the Federal Insecticide, Fungicide and Rodenticide Act ("FIFRA"), 7 U.S.C. § 136. To be used lawfully in the United States, a pesticide must be registered for a particular use by EPA under FIFRA.
Another statute, the Federal Food Drug and Cosmetic Act ("FFDCA"), 21 U.S.C. §§ 321-338, regulates food safety. Under the FFDCA, EPA establishes tolerances that authorize and place legal limits on the presence of pesticide residues on food. 21 U.S.C. § 346a. EPA must ensure that tolerances are set at levels that are "safe," as that term is defined in the statute. Id. Residues of a pesticide on food that exceed the levels permitted under a tolerance or for which there is no tolerance are deemed unsafe. Id. A commodity containing such unsafe residues is characterized as adulterated and is unlawful under the FFDCA. Id. § 342.
The Food Quality Protection Act ("FQPA"), enacted in August 1996, amended the FFDCA. In particular, FQPA defines the term "safe" and spells out several mandatory determinations that must be made for a tolerance to be granted. Under the FQPA safety standard, EPA can issue a tolerance only if it has determined that:
there is a reasonable certainty that no harm will result from aggregate exposure to the pesticide chemical residue, including all dietary exposures and all other exposures for which there is reliable information.
21 U.S.C. § 346a(b)(2)(A)(ii).
For carcinogens, Congress endorsed EPA’s policy of allowing no more than one additional cancer in one million exposed people from each pesticide use based on quantitative risk assessment using conservative assumptions. H. Rep. No. 104-669, part 2, 104th Cong., 2d Sess. at 40-41 (1996). Any change in this policy must be adopted as a regulation and must be scientifically based and at least equally protective of public health. Id.
In addition to adopting this new overall food safety standard, FQPA responded to evidence that existing food tolerances failed to account for the particular risks presented to children. National Academy of Sciences, Pesticides in the Diets of Infants and Children (1993). More specifically, EPA did not previously assess the particular toxicity of pesticides to infants and children or children’s unique exposures to pesticides and the risks presented by them. Id., Executive Summary at 3-7.
Reacting to this alarming evidence that children face inordinate risks from pesticide residues that had previously escaped analysis and regulation, FQPA mandates that EPA base its safety determination on an assessment of: (1) children’s consumption patterns that are likely to result in disproportionately high consumption of foods bearing the pesticide residue; (2) special susceptibility of infants and children to the pesticide residue, as well as the effects of in utero exposures; and (3) available information concerning cumulative effects on infants and children of cumulative exposures to other substances that have a common mechanism of toxicity. 21 U.S.C. § 346a(b)(2)(C)(i).
FQPA does more than merely require consideration of a pesticide’s potential effects on infants and children. It requires that, for threshold effects, "an additional tenfold margin of safety for the pesticide chemical residue and other sources of exposure shall be applied for infants and children to take into account potential pre- and post-natal toxicity and completeness of the data with respect to exposure and toxicity to infants and children." Id. § 346a(b)(2)(C). EPA may use a different margin of safety only if, based on reliable data, that margin will be safe for infants and children. Id.
FQPA also directs the Administrator to consider other factors, including aggregate exposures to the pesticide chemical at issue from a variety of food and non-food sources, cumulative exposures to pesticide chemicals that have a common mechanism of toxicity, and "whether the pesticide chemical may have an effect in humans that is similar to an effect produced by a naturally occurring estrogen or other endocrine effects." Id. § 346a(b)(2)(D)(v), (vi), (viii).
II. BASIS FOR THE OBJECTIONS
A. EPA Failed To Use A Tenfold Safety Factor To Protect Infants And Children.
1. EPA’s Stated Reasons for Declining to Apply FQPA’s Tenfold Safety Factor Are Specious.
In issuing the tolerance for vinclozolin on snap beans, EPA dispensed entirely with FQPA’s additional tenfold safety factor designed to protect infants and children. EPA offered two reasons.
First, EPA asserts that the 100-fold safety factor already provides adequate protection for the particular risks to children. EPA contends that the 100-fold safety factor takes into account two phenomena: "that people may be up to 10 times more sensitive to a pesticide than test animals, and that one person or subgroup of the populations (such as infants and children) could be up to 10 times more sensitive to a pesticide than another." 62 Fed. Reg. at 38,467. In EPA’s view, this latter component of the 100-fold safety factor provides children whatever added protection they need due to their special sensitivities.
Second, EPA limits the tenfold uncertainty factor to uncertainty or incompleteness in the toxicity data. Because EPA claims to have a complete data base for vinclozolin under existing guidelines, and the data reliably demonstrate fetal and post-natal toxicity, EPA concluded that the standard 100-fold safety factor would be safe for children, without the added margin of safety prescribed in FQPA. 62 Fed. Reg. at 38,466, 38,471.
This interpretation conflicts squarely with the plain language of FQPA, its legislative history, and EPA’s own description of the findings that it must make to avoid providing the additional safety factor for children.
Under FQPA, "an additional tenfold margin of safety for the pesticide chemical residue and other sources of exposure shall be applied for infants and children to take into account potential pre- and post-natal toxicity and completeness of the data with respect to exposure and toxicity to infants and children." 21 U.S.C. § 346a(b)(2)(C)(emphasis added). Since FQPA mandates the use of this tenfold safety factor for children in addition to the standard 100-fold safety factor, Congress obviously did not
believe that the 100-fold safety factor provides children adequate protection by itself.
Moreover, FQPA explicitly directs EPA to use the additional tenfold safety factor to take into account three situations with respect to infants and children: (1) pre- and post-natal toxicity; (2) incomplete exposure data; and (3) incomplete toxicity data. 21 U.S.C. § 346a(b)(2)(C); see also H. Rep. No. 104-669, pt. 2, at 43 (Administrator "is to apply" additional tenfold margin of safety "[w]hen data relating to infants and children are incomplete, and also to account for potential pre- and post-natal toxicity"). The fact that toxicity data may be complete for a particular pesticide does not eliminate the need for the tenfold safety factor to account for demonstrated fetal or post-natal toxicity or incomplete exposure data.
Indeed, the House Report clarifies that the FQPA tenfold safety factor is an extension of a tenfold factor that was previously applied by EPA and FDA to account for fetal developmental toxicity. Id. FQPA extends this existing safety factor to cover not only demonstrated fetal toxicity, but also post-natal developmental effects and incomplete toxicity and exposure data. Id.
Both FQPA’s plain language and its legislative history make clear that the tenfold margin of safety is designed to account for more than the completeness of the toxicity database under existing guidelines. Indeed, in its vinclozolin tolerance decision, EPA acknowledges that it may dispense with the additional tenfold margin of safety only "when EPA has a complete data base under existing guidelines and when the severity of the effect in infants or children or the potency or unusual toxic properties of a compound do not raise concerns regarding the adequacy of the standard margin/factor." 62 Fed. Reg. at 38,471 (emphasis added). Thus, EPA seems to recognize that a complete database, by itself, does not eliminate the need for the additional safety factor. Nonetheless, it casts off the additional safety factor in its vinclozolin tolerance decision on this basis alone.
2. The Tenfold Safety Factor Must Be Used In This Tolerance Decision.
Under the FQPA criteria, it is beyond question that the tenfold safety factor must be applied in this tolerance decision. First, the existing data reveal alarming fetal and post-natal toxicity effects. Vinclozolin is a demonstrated anti-androgen, causing sexual deformities in male offspring from fetal and post-natal exposures. Even under EPA and FDA’s pre-FQPA policy of using a tenfold safety factor to account for fetal toxicity, a tenfold safety factor is in order here. Additionally, under FQPA, which extends that tenfold safety factor to post-natal effects, the tenfold safety factor is clearly mandated. Vinclozolin’s demonstrated fetal and post-natal toxicity calls for greater, rather than less, protection for infants and children.
Second, EPA’s contention that it has a complete database on vinclozolin’s toxicity under existing guidelines does not justify its rejection of the tenfold safety factor. Of course, it is precisely the inadequacies in EPA’s existing data and assessments that spurred FQPA’s mandate to assess the special susceptibility of infants and children. Similarly, FQPA directs EPA to consider endocrine disruption effects and cumulative effects of substances with a common mechanism of toxicity -- two types of effects inadequately addressed in EPA’s existing guidelines. Given the new direction in FQPA, EPA’s existing guidelines cannot be the measure of whether toxicity data on a particular chemical are adequate and reliable.
Despite the solid body of scientific evidence of vinclozolin’s anti-androgen effects, the data on other toxic effects is spotty at best. The tolerance decision reveals little about vinclozolin’s neuro-behavioral effects. Other endocrine disrupting chemicals have been shown to affect sexual behavior at lower doses than those that cause the type of structural deformities observed in the vinclozolin studies. Guillette, Crain, Rooney & Pickford, "Organization versus Activation: The Role of Endocrine-disrupting Contaminants (EDCs) during Embryonic Development in Wildlife," 103 Environmental Health Perspectives 157 (1995); Mably, Moore, Goy & Peterson, "In Utero and Lactational Exposure of Male Rats to 2,3,7,8-Tetrachlorodibenzo-p-dioxin," 114 Toxicology & Applied Pharmacology 108 (1992); Tilson, Jacobson & Rogan, "Polychlorinated Biphenyls and the Developing Nervous System: Cross-Species Comparison," 12 Neurotoxicity & Teratology 239 (1990).
The vinclozolin tolerance decision acknowledges that EPA is in the process of developing a screening program for endocrine disruptor effects. 62 Fed. Reg. at 38,472. Until EPA has developed such a program and obtained the results of testing for neuro-behavioral effects, the toxicity database on vinclozolin cannot be considered to be complete. EPA purports to apply a threshold analysis to vinclozolin, but, given our current state of knowledge, it is controversial and laden with uncertainties to purport to set a safe level for exposures to a demonstrated endocrine disruptor.
The uncertainties in the toxicity data on neuro-behavioral and other endocrine disruption effects mandate use of the tenfold safety factor. The importance of this safety factor is elevated given EPA’s abandonment of other conservative assumptions in its tolerance decision.
As discussed below, EPA did not incorporate cumulative exposures to iprodione and procymidone into its analysis of threshold effects. EPA also ignored turf exposures to vinclozolin. In addition, EPA tolerated a cancer risk that is 2 and 1/2 times the level it deems to be negligible because of uncertainties in the risk assessment process. Finally, EPA estimated actual vinclozolin residues, thereby abandoning the conservative assumption that all crops will be treated.
Third, the FQPA tenfold safety factor is designed to account for incompleteness in exposure data. EPA ignores this purpose of the tenfold safety factor. In its tolerance decision, EPA fails to focus specifically and adequately on exposure data for nursing and non-nursing infants, despite vinclozolin’s demonstrated post-natal toxicity. Nor does EPA examine in sufficient detail the exposure patterns of other age children.
Instead, EPA purports to use whatever existing data it has based on the pre-FQPA database. Of course, the existing data sets are the very ones found to be inadequate by the National Academy of Sciences for assessing the impact of pesticides on infants and children. See National Acadamy of Sciences, Pesticides in the Diets of Infants and Children (1993). Specifically, the NAS found that "[b]oth government and industry data on residue concentrations in foods reflect the current regulatory emphasis on average adult consumption patterns," and that "foods eaten by infants and children are underrepresented." Id. at 5. "At present, data are derived from relatively small samples and broad age groupings, making it difficult to draw conclusions about the food consumption patterns of infants and children." Id. at 10; see also id. at 198. This problem is compounded by inadequacies found by NAS in the existing data on the levels of pesticides found on the foods consumed by infants and children. Id. at 211, 260-61, 314. It was in part to make up for these deficiencies that NAS proposed an additional tenfold safety factor.
FQPA directs the federal government to conduct surveys to document the dietary exposures of infants and children to pesticides. 21 U.S.C. § 346a(b)(2)(C). These surveys are designed to provide a more complete and adequate database for assessing children’s particular consumption patterns, but the results will not be available for some time.
Moreover, EPA failed to examine children’s exposures to pesticides as a result of non-food uses. While certain vinclozolin uses on turf will be discontinued, children may be exposed to vinclozolin as a result of its application for food crops. Given that irrigators in Oregon have been given no notice of vinclozolin applications, they have not been using protective gear, and vinclozolin residues may be brought into their homes on
their clothes, thereby exposing their children to the residues. EPA will need to assess such exposures in setting tolerances.
The exposure database for vinclozolin is incomplete at this time. It will not be complete until EPA has amassed an adequate database on the exposures of infants and children of various ages to the full range of foods in their diet and to pesticide residues from non-food uses.
EPA erred by not using the tenfold safety factor given vinclozolin’s severe fetal and post-natal toxicity, the inadequate toxicity data on vinclozolin’s neuro-behavioral effects, and the incomplete exposure data for infants and children.
B. EPA Failed To Incorporate Into Its Assessment Of Threshold Effects Cumulative Exposures To Pesticide Chemicals With A Common Mechanism Of Toxicity.
FQPA contains two mandates to incorporate cumulative effects into tolerance-setting decisions. First, EPA must assess a pesticide’s risks based on "available information concerning the cumulative effects on infants and children of such residues and other substances that have a common mechanism of toxicity." 21 U.S.C. § 346a(b)(2)(C)(i)(III). Second, EPA must consider such
information more generally apart from the particular impacts on children. Id. § 346a(b)(2)(D)(v).
In its vinclozolin tolerance decision, EPA acknowledges that vinclozolin, iprodione, and procymidone share similar toxicological and structural properties, and that they produce a common metabolite that has been associated with vinclozolin’s anti-androgen effects. 62 Fed. Reg. at 38,466. EPA incorporated exposures to iprodione and procymidone into its cancer risk assessment. Yet in assessing vinclozolin’s fetal and post-natal toxic effects, EPA inexplicably looked only at vinclozolin residues, ignoring cumulative effects that may be caused by exposures to iprodione and procymidone.
EPA tries to justify its failure to consider cumulative effects by claiming that it has not yet developed policies and methodologies for assessing common mechanisms of toxicity. In EPA’s view, "available information" includes agency policies and methods and thus FQPA’s direction to use "available information" on cumulative effects allows the agency to avoid a cumulative effects analysis before such time as it has developed related policies and methods. 62 Fed. Reg. at 38,470.
Such an interpretation is belied by the statutory language. FQPA directs EPA to use whatever information is currently available and to use additional safety factors to account for uncertainties in that information. This direction is designed to compel the agency to act on the basis of the data and information in hand, rather than wait until the agency has perfected its policies and methods for analyzing that data. If EPA is permitted to avoid considering available information prior to its development of new policies and methods, it will unilaterally be extending FQPA’s effective date indefinitely. The statute gives the agency no such power. To the contrary, statutory timelines mandate expeditious implementation of the FQPA.
The question must, therefore, be whether the agency has data available that enable it to assess cumulative effects. In the vinclozolin tolerance decision, EPA acknowledges that even now, in the absence of fully developed policies and methodologies, it may have sufficient data to make a common mechanism of toxicity determination for some chemicals. As EPA explains, some chemicals are clearly toxicologically dissimilar, thereby making it unlikely that they share a common mechanism of toxicity. Conversely, and of critical importance here, EPA concedes that pesticides that produce a common metabolite must be assumed to have a common mechanism of toxicity. 62 Fed. Reg. at 38,470.
In an outright contradiction, EPA concedes that vinclozolin, iprodione, and procymidone produce a common metabolite, yet EPA refuses to adhere to its previous statement equating that phenomenon with a common mechanism of toxicity. Id. at 38,470. Instead, the tolerance decision states that EPA has assumed that no common mechanism of toxicity exists. Id. at 38,466.
Despite this stated assumption, EPA included the cumulative exposures to vinclozolin, iprodione, and procymidone in its cancer risk assessment, but not in its assessment of other effects. There is absolutely no basis for EPA’s failure to incorporate cumulative effects into its threshold analysis.
* * *
These deficiencies are not merely technical. Had EPA used a tenfold margin of safety to account for vinclozolin’s risks to children, it could not have granted a tolerance for use of vinclozolin on snap beans. FQPA’s food safety standard would not be met.
By way of example, the tolerance decision contends that the public may be exposed to 14%-15% of the reference dose for chronic dietary risk. 62 Fed. Reg. at 38,472. Utilizing an additional tenfold safety factor would result in those same exposures exceeding the reference dose by 40%-50%. Similarly, chronic aggregate risks would increase from 13%-16% of the reference dose to 130%-160%.
In addition, the tolerance decision reports that the aggregate acute risk from dietary and drinking water exposures for females ages thirteen or more results in a margin of exposure of 380. EPA considers the margin of exposure to be of concern when it is less than 100. When adjusted by the tenfold safety factor, the margin of exposure for this risk would be 38. See also id. (short- and intermediate-term aggregate risk would be reduced from a 140-150 margin of exposure to 14-15).
These risk estimates are based on vinclozolin alone, without considering iprodione and procymidone. Id.
Accordingly, NRDC asks EPA to rescind the tolerance granted for vinclozolin and to conduct a review and initiate revocation proceedings, as necessary, of other tolerances for vinclozolin, iprodione, and procymidone to bring the tolerances in line with FQPA’s standards.
FACTUAL ISSUES ON WHICH EVIDENTIARY HEARING IS REQUESTED
I. EPA SHOULD GRANT AN EVIDENTIARY HEARING, PRODUCE ITS ADMINISTRATIVE RECORD, AND TAILOR FURTHER PROCEEDINGS TO THE FACTUAL ISSUES THAT REMAIN AT ISSUE.
This tolerance-setting process has been plagued by inadequate disclosures to the interested public. In its initial notice, EPA simply published BASF Corporation’s summary of its tolerance petition, which relied on internal agency documents that are not available to the public. EPA also failed to notify the public of its position on various controversial aspects of the petition or matters on which the petition provided insufficient information to enable EPA to make the statutorily mandated inquiries and findings.
In April, 1997, the submitters filed a FOIA request seeking the full record on which the tolerance-setting process is based. However, EPA has produced no documents in response to this request.
NRDC objected to the inadequate public disclosure provided by BASF’s summary. In responding to this comment, EPA relied, in part, on the objection process, claiming that the public would have an opportunity to comment more fully on the actual agency decision at that time. 62 Fed. Reg. at 38,466.
Of course, that opportunity to comment comes in the form of objections to the final agency decision. The public then knows the basis for the agency’s decision only because the final decision has been made. Once the tolerance has been granted, residues of the pesticide are permitted on food to the detriment of the affected public.
In any event, in this instance, the tolerance decision fails to provide a sufficient explanation of the basis for various aspects of the agency decision. For example, the tolerance decision does not set forth the consumption data available to and/or used by EPA to determine the exposures of infants and children to vinclozolin. Accordingly, the objection process will enable the public to provide meaningful and fully informed input only if an evidentiary hearing or some analogous process proceeds.
As set forth in EPA’s regulations, the objection process could proceed without ensuring that the objector has access to the information on which the agency based its decision. The regulations appear to be designed to facilitate review by the company that petitioned for the tolerance, rather than the general public. Indeed, the fee schedule is in line with the economic wherewithal of those with a financial stake in the tolerance. A company filing objections would generally be privy to EPA’s full record on the pesticide and often to the agency’s proposed action, as evidenced by BASF Corporation’s access to EPA’s draft registration eligibility document for vinclozolin, as described in BASF Corporation’s summary of the tolerance petition. In contrast, the public may have had little previous access to the agency’s records and review of this particular pesticide.
EPA’s regulations appear to allow EPA to resolve objections without holding an evidentiary hearing or otherwise making the administrative record available to the objectors. In contrast, under notice and comment rulemaking, the public must be provided ample notice of the proposed agency decision and an opportunity to submit comments, which must be considered by the agency before it makes its final decision. Moreover, in any judicial challenge to the final agency regulation, a complaint must be filed based on the final regulation, but judicial review is based on the full administrative record produced by the agency and any additional information admitted by the court to supplement that record. See, e.g., Southwest Center for Biological Diversity v. U.S. Forest Service, 100 F.3d 1443, 1450 (9th Cir. 1996); Love v. Thomas, 585 F.2d 1347, 1357 (9th Cir. 1988).
Under EPA’s regulations, NRDC is required to describe, at this time and on the basis of the information now available to them, the issues on which they request an evidentiary hearing and the basis for their objections. However, until they have obtained access to the administrative record for the tolerance decision, the submitters cannot accurately identify the issues on which that record is insufficient. It may be that some of the issues on which the record is now incomplete were addressed in documents that were before the agency during this administrative process.
NRDC urges the agency to provide the full record on which EPA based its tolerance decision and to give the submitters an opportunity to supplement their hearing request and objections once they have reviewed that record. Some of the issues identified in this request for an evidentiary hearing may, in fact, be resolved based on an examination of the full record and may not need further development at a hearing. Moreover, any remaining factual issues may be further narrowed and refined in light of the evidence in the record.
If EPA resolves NRDC’s objections without first producing the administrative record, the next stage for review is the court of appeals. If NRDC seeks such review, it could be cumbersome to address inadequacies in the administrative record for the first time in the court of appeals. It would streamline the review
process and even the playing field for EPA to produce the record before this objection process proceeds any further.
If EPA is unwilling to make this accommodation, then it must grant NRDC’s request for an evidentiary hearing. Otherwise, NRDC would be denied any opportunity to obtain the record and supplement it, where necessary.
II. STATEMENT OF FACTUAL ISSUES ON WHICH AN EVIDENTIARY HEARING IS SOUGHT
Pursuant to 40 C.F.R. § 178.27, NRDC seeks an evidentiary hearing on the following issues. NRDC is not providing copies of documents because it does not yet know which documentation EPA relied upon in making its tolerance decision. In any event, the matters on which NRDC seeks an evidentiary hearing focus on the information available to the agency during the course of this tolerance-setting process. Accordingly, this information is in the hands of the agency.
Upon reviewing the information before the agency, NRDC may wish to submit evidence to show that EPA failed to consider critical information and that it had a legal obligation to consider such information in making its tolerance decision. NRDC cannot identify the evidence that it would submit until it has reviewed the full record before the agency. Nonetheless, NRDC’s objections describe the nature of the evidence upon which it relies in this challenge.
A. Exposure Data.
What data are available concerning the exposure of infants and children to vinclozolin?
Whether EPA’s exposure estimates take into account the particular exposure patterns (including food and non-food) of infants and children?
How long may existing stocks of vinclozolin be used for uses that may soon be deleted from the label?
These issues relate to the first objection: that EPA failed to use a tenfold safety factor to protect infants and children. The FQPA mandates the use of this safety factor in part in order to account for uncertainties in data regarding the exposure patterns of infants and children. Thus, unless EPA can demonstrate that it based its risk assessment on complete data regarding the specific consumption and exposure patterns of infants and children of different ages, its tolerance violates the FQPA by failing to apply the tenfold safety factor, and the relief requested by the submitters must be granted. That is, EPA must immediately rescind the tolerance already issued for vinclozolin residues on snap beans and revise its risk assessment to incorporate the tenfold safety factor. Such a revision will require EPA to reject BASF’s petition because in many, if not all, of the risk calculations it will cause the exposure level to exceed the reference dose or cause the margin of exposure to drop below 100.
B. Incompleteness of Toxicity Data.
What toxicity data are available concerning vinclozolin’s toxic effects, including data on the particular sensitivities and susceptibilities of infants and children and vinclozolin’s endocrine disruption effects?
What data did EPA use on vinclozolin’s toxicity, including data on the particular sensitivities and susceptibilities of infants and children to vinclozolin?
These issues also relate to the first objection: that EPA failed to use a tenfold safety factor to protect infants and children. The FQPA mandates the use of this safety factor not only to account for uncertainties in data regarding exposure but also to account for uncertainties regarding toxicity of the pesticide residue to infants and children. Thus, unless EPA can demonstrate that it based its risk assessment on complete data regarding the specific toxicity of vinclozolin to infants and children -- and that these data are complete under the standards of the FQPA, not simply based on "existing guidelines" -- its tolerance violates the FQPA by failing to apply the tenfold safety factor. As above, the relief requested by the submitter is that EPA immediately rescind the tolerance already issued for vinclozolin residues on snap beans and revise its risk assessment to incorporate the tenfold safety factor.
C. Available Data on Common Mechanism of Toxicity.
What data are available on the similarities in the toxicological and structural properties of vinclozolin, iprodione, and procymidone, including the production of a common metabolite?
Whether available data permit a cumulative effects assessment of vinclozolin, iprodione, and procymidone, particularly in connection with the fetal and post-natal effects?
Why did EPA exclude California in its assessment of exposures to vinclozolin in drinking water resulting from application on strawberries?
These issues relate to the second objection: that EPA failed to incorporate into its assessment of threshold effects cumulative exposures to pesticide chemicals with a common mechanism of toxicity. The FQPA requires EPA to consider "available information concerning the cumulative effects of [the pesticide residue at issue] and other substances that have a common mechanism of toxicity." 21 U.S.C. 346a(b)(2)(D)(v). EPA defends its failure to consider cumulative effects (except with respect to cancer risk) on the grounds that sufficient information is not "available." Accordingly it is necessary to determine what information relevant to this consideration is available and whether such information and data are sufficient to support an analysis of the cumulative effects of vinclozolin and related compounds. If it is, then EPA has violated the FQPA with respect to this requirement as well, and the relief requested by the submitters -- that EPA immediately rescind its tolerance for vinclozolin on snap beans, revise its assessment of non-cancer risks to account for cumulative effects, and conduct a review and initiate revocation proceedings, as necessary, of other
tolerances for vinclozolin, iprodione, and procymidone to bring the tolerances in line with FQPA’s standards -- must be granted.
Respectfully submitted,
Patti Goldman
Amy Sinden
Earthjustice Legal Defense Fund
705 2nd Avenue, Suite 203
Seattle, WA 98104
(206) 343-7340
Al Meyerhoff
Natural Resources Defense Council
71 Stevenson Street, Suite 1825
San Francisco, CA 94105
(415) 777-0220
September 15, 1997
11/15/97