730 P
§ t±~'\. UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
~ ^ ****'- "> . WASHINGTON, D.C. 20460
12/8/95 OFFICE OF
PREVENTION. PESTICIDES. AND
' ' TOXIC SUBSTANCES
TO WHOM IT MAY CONCERN:
The attached draft PR Notice entitled "Toxicologically Significant Levels of Pesticide
Active Ingredients" is being made available for public comment. The notice focuses on
cross-contamination of active ingredients in pesticide products.
A Notice of Availability of this draft notice will be published shortly in the Federal
Register which will invite public comment for a period of sixty days. If you have any
comments, you may cite the Docket Number OPP-00424 and send them to:
1
By U'.S. Mail: ''; V , .
Public Docket and Freedom of Information Section
Field Operations Division ,
Office of Pesticide Programs
U.S. Environmental Protection Agency (7506C)
Washington, D.C. 20460 ' - ; ' ' , > .
Hand Delivered: > . ' - -
Public Docket and Freedom of Information Section .
Field Operations Division .
Office of Pesticide Programs
Room 1132, Crystal Mall 2
1921 Jefferson Davis Highway
Arlington. VA 22202 " ,
If you have any questions about the implementation of this policy, you may contact
Jim Jones at 703-3Q8-8358. If you have questions regarding the analysies supporting this
policy, contact Nancy Fitz at 703-305-7385.;; ' ' , '-
Sincerely, , ' V..
Stephen L. Johnson, Director
Registration Division
Recycled/Recyclable . f'nmea win, Vegelable Oil Based InKs on 100'<, Recyclecl Paper (40% Postconsumer)
-------
-------
*** 12/8/95 DRAFT***
PESTICIDE REGULATION (PR) NOTICE 96-? . '
NOTICE TO MANUFACTURERS, FORMULATORS, PRODUCERS
1 AND REGISTRANTS OF PESTICIDE'PRODUCTS
.ATTENTION: , Persons Responsible for Registration of Pesticide
Products ." - '
SUBJECT: Toxicologically Significant Levels; of Pesticide
Active Ingredients /
This notice sets out the Environmental Protection Agency's
(EPA's) new risk-based approach ,for reporting "toxicologically
significant" contaminants that are also pesticide active
ingredients (AIs). The Notice establishes concentration levels
of such contaminants ,that ,will generally be considered
toxicologically significant in specified types of pesticide
products., This draft Notice is subject to a 60 d.ay comment
period. . : - -
' I. BACKGROUND . ,'....', ..".':.-
40 CFR 158'. 167 requires all impurities to be reported and
'approved/accepted by the Agency as part of the product's . ,
registration prior to the sale or distribution of the product
cpntaining the impurities in their products that are of "
toxicological significance. The.introductory.paragraph to
§ 158.167 states, in part, that "... If the applicant has reason
to believe that an impurity that EPA would consider
toxicologically significant may be present, the discussion must
include an expanded discussion of the possible formation ^of the
impurity and the amounts at which .it might be present. "
Reporting is required for all impurities meeting this standard
regardless of where the contamination occurred in the production
and distribution processes. ' ;
EPA also requires applicants a.nd . registrants to propose
certified limits (legally permissible, amounts) for certain
toxicologically significant impurities in pesticide products. If
the product is a technical grade of active ingredient or is
produced by an integrated system, 40 CFR 158.175(a)(3) requires
that the applicant provide-a proposed certified upper limit (the
"maximum legally permissible amount) for each impurity of
toxicological significance associated with the active ingredient
and found to be present in any sample of the product. In,
addition, EPA may, on a case-by-case basis, require certified
-------
limits for other toxicologically significant impurities reported
under § 158.167. Certified limits are set forth on the
applicant's confidential statement of formula (CSF).
The preamble to the May 4, 1988 rule that promulgated 40 CFR
158.167 and 158.175 provided two sets of criteria for determining
impurities having potential toxicological significance. The
first was a list of specific substances known to be of
toxicological significance. Among these categories, the Agency
included "Any impurity that is also an active ingredient." (53 FR
15973) The Agency did not provide a numeric standard for
determining the point at which these substances are considered
"toxicologically significant." Rather, EPA has taken the
position that any level of a chemical which is identified by the
Agency as an active ingredient that is a contaminant in a product
is of potential toxicological significance and must be reported
to EPA or identified on the CSF..1
If an AI is detected that is not on the CSF or that has not
been reported as described above, then the registrant has
violated section 12(a)(l)(C) of the Federal Insecticide,
Fungicide, and Rodenticide Act (FIFRA) because the registrant has
distributed or sold a "registered pesticide the composition of
which differs at the time of its distribution or sale from its
composition as described in the statement required in connection
with its registration under [FIFRA] section 3."
The policy announced in this notice refines the category of
the existing definition of "toxicologically significant"
regarding impurities that are active ingredients in other
products by setting risk-based levels at which such substances
will generally be considered "toxicologically significant." This
policy does not address or alter any other portion of the
existing definition.
Since 1992, several EPA Regions have conducted
investigations to evaluate the extent of cross contamination of.
bulk pesticides. An important part of applying the 40 CFR Part
158 standards to bulk pesticides at repackaging/refilling
establishments (often retail dealers) is EPA's interpretation of
the Bulk Pesticides Enforcement Policy (Bulk Policy) dated July
11, 1977 and amended on March 4, 1991. Nothing in this policy
changes the conditions outlined in the Bulk Policy for the two
parties involved (registrant and repackager.) In particular,
EPA's position that both parties are accountable for the
integrity of the product as set out in the Bulk Policy remains
the same. For example, even when the contamination occurs at the
bulk repackaging facility both parties remain responsible for the
integrity of the product.
As a result of many questions and comments from EPA Regions,
States and the regulated industry, the Office of Pesticide
-------
Programs has re-evaluated EPA's current position on cross
contamination as described above. As a result of this work, EPA
is revising its interpretation of "toxicoiogically significant"
levels of AIs to be the risk-based.levels set out in section IV.
However, this policy.only addresses impurities that are also AIs,
i.e., a subset of the universe of'impurities. Specifically, for
the purposes of this policy, a contaminant is defined as an AI
that is not on the product's confidential statement of formula or
listed in the discussion of impurities. EPA's position on other
impurities has not changed.
II. OBJECTIVES . ' ''' . '.' - - ' :
EPA determined that a policy on cross contamination should:
Ensure that allowable cross contamination does not pose
unreasonable adverse effects;
Recognize that cross .contamination is a reality, and that
not all cross1contamination is problematical;
Set a clear standard that can be readily applied by
.EPA/States and the regulated industry alike;
, Be easy to administer in that it minimizes the paperwork
burden for EPA and registrants;
Maintain accountability for the product from the registrant
to the end user; ,
Minimize the possibility that cross contamination will pose
a significant problem, but ,
Allow for easy and equitable enforcement when
toxicoiogically significant cross contamination is found,
and " v. _ -.-./ -.,.''.' . '
Not preclude marketplace/private solutions to correct
problems that,do arise. :
III. APPROACH . '' ; ..'' . -. . . ; '.
EPA decided that a risk-based approach would be most likely
to meet these objectives and to define an acceptable policy, "one
risk-based option would be to consider each case of cross
contamination individually after it had been detected, to
evaluate the potential risks (e.g., to humans, birds, aquatic
organisms, plants, etc.) and to determine if any of EPA's levels
of concern were exceeded. However, EPA rejected this option
because it would be too resource-intensive and would address this
issue after harm occurred as opposed to preventing the potential
f or harm. ', ' . . ' .-.;
Instead, EPA considered the risks for several endpoints,
including human health, violative residues, contamination of
ground water and ecological effects to determine which endpoints
would be most sensitive to cross contamination and what levels of
cross contamination could be tolerated and remain .generally
protective of human health and the environment. For each
-------
endpoint, an analysis was done to evaluate a reasonable worst
case scenario or a range of potential scenarios to see if an
overall, generally protective contaminant concentration could be
determined. EPA grouped contaminants and pesticides into
different categories (see the table in section IV), to yield a
scheme of toxicologically significant concentrations.
The following is a brief description of the end points that
were considered. In most cases, it appears that phytotoxicity to
the target plants is the most sensitive endpoint and, therefore,
the limiting factor in determining toxicological significance.
Human health effects. Because cross contamination is most
likely an intermittent event, short-term exposure is most likely.
Therefore, EPA believes that,the potential for chronic human
health effects from cross contamination is negligible. EPA
focused on the potential risks to individuals who would be
handling contaminated products. The analyses of these human
health risks show that acute risks to humans at the cross
contamination levels allowed by this interpretation are also
negligible. EPA also considered contamination in pesticides
applied to the human body.
Violative residues. EPA evaluated the potential for cross
contamination to result in violative residues in food or feed.
Theoretically, it is possible that a contaminant could be
detected in food or feed and for there to be no tolerance for the
contaminant pesticide. EPA's analysis indicates that this is a
highly unlikely scenario. In addition, because cross
contamination occurs intermittently and would be at low levels,
EPA believes that dietary risk from the potential residues
allowed by this interpretation 'would be negligible.
Ground water. The possibility of the contamination of
ground water was raised as a potential concern in locations with
sandy soils and shallow aquifers. The Florida Department of
Agriculture and Consumer Services (DACS) conducted a preliminary
ground water modeling exercise using a number of conservative
assumptions regarding leachability, pesticide half-life, and
product application rate. The Florida DACS concluded that, while
contamination of ground water was possible, it was of minimal
concern because pesticide AIs at the; levels allowed by-this
notice are unlikely to move to ground water in concentrations
that would pose significant risk to human health.
Ecological effeets/phytotoxicity. Based on a preliminary
review of potential ecological effects from cross contamination
(e.g., risks to birds, aquatic organisms and plants), EPA
believes that plant toxicity, or phytotoxicity, is the most
sensitive endpoint given the relatively low concentrations of
pesticides being considered and poses the greatest potential for
harm. The phytotoxicity analyses described below focused on the
-------
direct application of the contaminated product to terrestrial
plants because this scenario represents a higher level of ;
exposure than other exposure pathways, such as runoff and drift.
IV. TOXICOLOGICALLY SIGNIFICANT LEVELS OF CONTAMINATION
The following table defines the levels of contaminants that
EPA considers to be toxicologically significant. Specifically,
the presence of a contaminant at a concentration greater than the
concentration specified in the table would generally be
considered toxicologically significant. Each contaminant should
be considered individually. The basis for each category is
described in section V below.
Toxicologically Significant Levels of Contaminants
Category
1 ' .
2
3
4
. . 5
6
7
8
9 .
Scenario
Insecticide :, rodenticide or
fungicide in any insecticide,
fungicide or herbicide
Herbicide in any pesticide where
the contaminant ' is accepted for
use on all crops for which the
product is labeled.
Any pesticide other than a low
application rate- herbicide in an
antimicrobial pesticide
Standard herbicide * 'in any
herbicide
Standard herbicide in any
insecticide or .fungicide
Low application rate herbicide 5 in
a. low application rate herbicide
Low application rate herbicide in a
standard herbicide.
Any pesticide in a pesticide
applied to the human body
Low application rate herbicide in a
pesticide other than a herbicide
%
Toxicologically
Significant ( Level (ppm)
1000 .
1000
1000
250
' 100
Level of quantification
6 pr 100 ppm, whichever
is higher
Level of quantification
or 20. ppm, which ever
is higher
10
Level of quantification
or 1 jppm, whichever is
higher
1. This column presents the 'toxicol'ogical ly significant level, i.e.
concentration above which EPA would consider the contaminant to be
toxicologically significant. . :,
the
-------
2. The FIFRA definition of insecticides includes miticides.
3. For the purposes of this policy, a contaminant is defined as an AI that is
not on the product's confidential statement of formula (CSF) or listed in the
discussion of impurities. .
4. For the purposes of this proposal, a Standard herbicide is defined as a
herbicide with a maximum labeled application rate of AI on labeled crops
>0.5 pounds Al/acre.
5. For the purposes of this proposal, a low application rate herbicide is
defined as a herbicide with a maximum labeled application rate of AI on ,
labeled crops <0.5 pounds Al/acre. ,
6. For purposes of this notice, the level of quantification is the level of
quantification achievable by EPA at the time the analysis is performed.
In establishing levels of toxicological significance, EPA
realizes there is a trade-off among competing factors. A lower
standard would be more protective, but would lead to higher
compliance costs for the industry, higher implementation costs
for EPA and the industry (e.g., the submission of expanded
discussions of impurities in many cases instead of just a few),
and the potential for fewer products available for users. When
evaluating the results of the analyses discussed in section V,
EPA attempted to find a reasonable balance between these
competing factors. If future experience indicates that these
values are not sufficiently protective, the Agency may find it
appropriate to modify these levels of toxicological significance.
EPA believes the values in the table are generally
protective in most contaminant/product combinations. Because it
is impossible to consider every potential contaminant/product
permutation, adverse effects could occur when contamination is
present at or below the concentrations in the table,.
The Agency recognizes that the proposed standards will not
prevent all possible adverse effects from occurring; this is not
a zero risk standard. For example, EPA is aware of a situation
where a standard herbicide (norflurazon) contaminated an
insecticide (insecticidal soap for use on home and garden plants)
at levels below LOO ppm (as set out in Category 5) and plant
damage occurred. The Agency-;will continue to deal with such
situations using other regulatory tools including section 6(a)(2)
of FIFRA. , . ,
Accordingly, this policy doesnot excuse applicants or
registrants from the requirement to submit to EPA factual
information regarding unreasonable adverse effects of a pesticide
under section 6(a)(2) of FIFRA and EPA regulations at 40 CFR . '
152.50(f)(3). . If an applicant or registrant possesses factual
information not previously reported to EPA indicating that a
contaminant in a product may pose risk to human health or the
environment in concentrations lower than those specified in the
above table, that information must be submitted to EPA. Failure
to submit such information on a timely basis is a violation of
sections 12(a)(2)(B)(ii) and 12(a)(2)(N) of FIFRA. In addition,
-------
the distribution or sale of any product containing an unreported
contaminant for which the registrant has failed to submit
additional factual information regarding unreasonable adverse
. effects is a violation of section 12(a)(lj(C) (composition
differs)' of FIFRA.
v. DISCUSSION ' . ;.- ''.'''..: :
This section summarizes the analyses that support the
toxicologically significant levels in the above table. For more
details, refer to the full analyses (see section VIII). .EPA ;
believes these analyses represent reasonable worst case or
representative scenarios. Because it was impossible to consider
every potential contaminant/product permutation, however, the
-analyses are necessarily limited in scope. '
Category .1. Because phytotoxicity is generally not a major
concern with insecticides or fungicides, the limiting endpoint
>for this category is potential adverse effects to human health.
EPA conducted a risk assessment assuming a fungicide was
contaminated with iooo ppm of an insecticide, fungicide and
herbicide (each individually). ^Exposure to mixer/loaders and
applicators during airblast and aerial treatment was analyzed,
because these .application methods would likely result in the
greatest potential worker exposure. Based on this limited
analysis, EPA concludes that adverse effects to human health
occurring, from cross contamination at 1000 ppm are very unlikely
to occur. It is, however, theoretically possible that certain
..exposure scenarios may present risks which exceed EPA's levels of
concern (LOC's), e..g,, when highly acutely toxic pesticides
contaminate products that, reguire the use of limited personal
protective equipment. However, EPA believes that contamination
at or below 1000 ppm generally is not of toxicological concern
t'ror a human health perspective. (References,! and 2)
Kith the exception of pesticide products that are
intentionally applied to the human body (see category 8), other
user scenarios involving cross contamination were considered and
were found to pose negligible risk to human health. (Referenced)
EPA also evaluated the potential for illegal residues as a
result of cross contamination. The conditions that are most
likely to, result in.possible violative residues are when
application 'rates are high, pre-harvest intervals are short and
the harvested,crop can be expected to have high tolerance values.
However, actual residues on commodities rarely approach the legal
tolerances. An.examination of Food and Drug Administration
surveillance data for the example fungicide on several
conroditics shows that actual residues are usually far below the
tolerances (2',5C;, - ^700 times less in this case). Therefore, EPA
believes that it is hiahly unlikely that illegal residues, from
cross contamination at or below 1000 ppm would'be found. It is
-------
important to note, however, that food or feed containing residues
from cross contaminants may be adulterated pursuant to section
402 of the Federal Food, Drug, and Cosmetics Act (FFDCA) unless
there is a tolerance covering the residue. (References 1 and 2)
Category 2. The limit in category 2 is based on the
assumption that it is safe to apply 1000 ppm (0.1% by weight) of
a herbicide to a crop or site for which the herbicide is labeled
for use. The herbicide that is the contaminant in this case will
be present at a much higher concentration when it is an intended
AI in a pestici'de product used on the same crop or site.
Category 3. Similar to the analyses supporting category 1,
EPA considered an apparently worst case scenario contamination
of an antimicrobial pesticide with an insecticide (azinphos-
methyl) at 1000 ppm. The estimated maximum exposure was based on
a Chemical Manufacturers Association exposure assessment study of
a wide variety of use sites and application methods. The
exposure to the insecticide did not exceed EPA's LOG. Based on
this limited analysis, EPA concludes again that adverse effects
to human health occurring from cross contamination at 1000 ppm
are very unlikely. ,
Categories 4-9. The remaining five categories (with the
exception of category 8) involve herbicides as the contaminants
and the levels of toxicological significance are based on
phytotoxicity analyses. The categories are distinguished by
whether the contaminant is a standard or a low dose herbicide and
by the type of product that is contaminated. While the
distinction between standard and low dose herbicides is based on
the AI application rate, the two categories represent different
levels of plant toxicity. Low dose herbicides generally are
phytotoxic at lower levels than standard herbicides. Therefore,
these AIs can be applied at lower rates than standard herbicides.
A corollary is that, when a low dose herbicide is the
contaminant, a relatively low concentration is toxicologically
significant. The category of low dose herbicides generally
includes chemicals classified as sulfonylureas, imidazolinones
and triazolopyrimidines.
The type of product that is contaminated also affects the
level of contaminant that would be toxicologically significant.
In most cases, insecticides and fungicides are applied more often
and at higher rates than standard herbicides, which are applied
at higher rates than low dose herbicides. For example, a given
herbicide contaminant would be present in similar quantities on a
particular crop if it was present in a fungicide at a low
concentration and in a standard herbicide at a higher
concentration. . ,
Category 4. .Category 4 is based on an analysis of 25
i
8 "'....''
------- |