Attachment H: Response to Comments on the Public Review Draft of the ICR

Docket ID No. EPA-HQ-OPPT-2007-1081

Response to Comments
on the
Public Review Draft
of the
Information Collection Request (ICR) entitled:
“Tier 1 Screening of Certain Chemicals Under the
Endocrine Disruptor Screening Program (EDSP)”
EPA ICR No. 2249.01, OMB Control No. 2070–new
(72 FR 70839, December 13, 2007)
April 10, 2009

Table of Contents
Introduction ..................................................................................................................... 2
Commenters.................................................................................................................... 2
Comments and Responses ............................................................................................. 3
1.

Validation of Tier 1 and 2 Assays is a Prerequisite ............................................ 3

2.

Duplication of Data............................................................................................. 6

3.

Practical Utility.................................................................................................. 10

4.

Methodology for Estimating Burden and Costs ................................................ 11

5.

Estimated Test Costs ....................................................................................... 13

6.

Estimated Burden and Costs............................................................................ 15

7.

Small Entity Burdens........................................................................................ 15

8.

Other Topics..................................................................................................... 16

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

Introduction
On December 13, 2007, EPA issued two Federal Register (FR) notices seeking
public comment on the Endocrine Disruptor Screening Program (EDSP). The first FR
notice sought public comment on the draft policies and procedures that the Agency
intends to use for the initial screening of pesticide chemicals under the Agency's EDSP
(73 FR 70842, docket ID No. EPA-HQ-OPPT-2007-1080). The other sought public
comment on the draft Information Collection Request (ICR) that describes the
information collection activities associated with Tier 1 screening of the first group of
chemicals under the EDSP and provides EPA's estimates for the related paperwork
burden and costs (72 FR 70839, docket ID No. EPA-HQ-OPPT-2007-1081).
This document provides a summary of the comments received on the draft ICR,
and the Agency’s responses to those comments applicable to the ICR document. For
information about the Agency’s responses to the comments received about the
proposed policy and procedures, please go to docket ID No. EPA-HQ-OPPT-20071080. To view the comments received, please go to the docket for these documents at
www.regulations.gov.

Commenters
The following 11 entities filed substantive comments on the draft ICR (does not
include any commenters that simply asked for an extension of the comment period):
Ref (1)
04

Document ID # (2)
EPA- HQ-OPPT-2007-1081-0004

05

EPA- HQ-OPPT-2007-1081-0005

06

EPA- HQ-OPPT-2007-1081-0006

07

EPA- HQ-OPPT-2007-1081-0007

09

EPA- HQ-OPPT-2007-1081-0009

10

EPA- HQ-OPPT-2007-1081-0010

11
12

EPA- HQ-OPPT-2007-1081-0011
EPA- HQ-OPPT-2007-1081-0012

13

EPA- HQ-OPPT-2007-1081-0013

14

EPA- HQ-OPPT-2007-1081-0014

15

EPA- HQ-OPPT-2007-1081-0015

Commenter (3)
B. Sachau
Carla J. Mattingly, National Engineering Regulatory
Compliance Coordinator, Centennial Communications
Michael C. White, Director of Regulatory Affairs, Chemical
Producers and Distributors Association (CPDA)
Dee Ann Staats, Environmental Science Policy Leader,
CropLife America
Larry E. Hammond, Chairman, Technical Committee,
Industry Task Force II on 2,4-D Research Data
Michael P. Walls, Managing Director, Regulatory and
Technical Affairs, American Chemistry Council (ACC)
Kimberly S. Gilbert, Regulatory Leader, Dow AgroSciences
Scott Slaughter, Center for Regulatory Effectiveness (CRE)
Dee Ann Staats, Environmental Science Policy Leader,
CropLife America (2nd Comments)
Susan Ferenc, President, Chemical Producers and
Distributors Association (CPDA)
Beth L. Law, Assistant General Counsel, Consumer
Specialty Products Association (CSPA)

KEY:
(1) This is the number that is used in this document to refer to this particular commenter.
(2) This is the number that is used to identify this comment in the docket at www.regulations.gov
(3) This is the name of the individual or entity that submitted the comments, along with their affiliation, if
provided.

Page 2 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

These 11 commenters provided similar comments that can be grouped into
specific topic or subject categories. This document is organized according to the topic
or subject categories.

Comments and Responses
1. Validation of Tier 1 and 2 Assays is a Prerequisite
Comment: Commenter #12 stated that the record does not demonstrate this ICR’s
consistency with the Paperwork Reduction Act (PRA) requirements, 44 U.S.C. §§
3506(4)(b)(i)(c),(d); 44 U.S.C. § 3507(a). Specifically, the commenter asserted that
EPA cannot satisfy the PRA requirements until EPA has validated and published all of
the tier 1 and tier 2 assays being considered for the EDSP. Only then can the public
comment on-- and only then can the Office of Management and Budget (OMB)
determine-- the practical utility, accuracy and reliability, cost and burden of the EDSP.
They believe that this cannot be separated for the purpose of public comment and
compliance with the PRA. To explain their comment, they provided this example:
“[I]f a compound elicits a positive outcome in the tier 1 screening assays, then it
will have to perform tier 2 assays. The cost and burden of the tier 2 assays flow
from the tier 1 assay results, and the tier 1 assays are useless without the tier 2
assays, because as EPA stated, “the ultimate purpose of the EDSP is to provide
information to the agency that will allow the agency to evaluate the risks
associated with the use of a chemical and take appropriate steps to mitigate any
risks.” 72 Fed. Reg. 70842, 70844 (Dec. 13, 2007). Consequently, the cost,
burden and usefulness of the tier 1 assays depend on the tier 2 assays. EPA
itself has cautioned that the practical utility of the tier 1 assays cannot be
determined until their results can be compared to the tier 2 assay results: “while it
is of interest to know how well these [tier 1] screens perform in identifying
chemicals that are positive in tier ii tests, this can only be done to a limited extent
at this time. Examples of this type of assessment have been conducted with the
uterotrophic and hershberger in vivo screens against other in vivo data including
multi-generational tests. However, the real proof of the performance of the tier I
screens will be a retrospective comparison of the performance of the battery with
tier II results after sufficient tier II data have been generated in the testing
program. This is why EPA is committed to a retrospective analysis of the test
data generated on the first 50 to 100 chemicals tested in the EDSP.”
Since the EPA record does not show compliance with the PRA’s standards, the
Commenter states that EPA must withdraw this ICR until the agency believes it has
validated both the tier 1 and tier 2 assays. If and when EPA has validated both the tier
1 and tier 2 assays, then EPA should allow another public comment period on a
proposed ICR before EPA sends an ICR to OMB for review.
In addition, the Commenter stated that the EPA record doesn’t demonstrate this
ICR’s compliance with EPA’s guidelines under the Information Quality Act (“IQA”), 44

Page 3 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

U.S.C. § 35161, because EPA has not demonstrated that the tests generate accurate
and reliable results.
EPA Response: The Agency disagrees with this commenter. The ICR is in
compliance with the PRA and EPA’s IQGs.
1. The scope of ICR was stated as being the information collection activities related
to Tier 1 screening only.
First and foremost, it is important to remember that this ICR only applies to the
Tier 1 assays. Any discussion in the ICR of Tier 2 under the EDSP is provided for the
sole purpose of describing the overall program, and was not provided for the purposes
of obtaining approval for any potential future Tier 2 activities under the PRA. Any
activities related to the Tier 2 testing phase of the EDSP will be addressed in a future
ICR. Although the Tier 2 assay validation process is underway, the Agency has stated
that it will address the paperwork activities related to the as yet undefined procedures
for requiring Tier 2 testing in a future ICR. There are NO current Tier 2 related
paperwork activities that require approval under the PRA. This was specifically
explained in the draft ICR, as well as in the related draft procedures document.
To clarify this further, the Agency has revised the title of the ICR to specifically
reflect the “Tier1” focus, and has revised the discussion to make this scope very clear.
2. Completion of the validation process for all of the assays in the final Tier 1
battery is not a prerequisite for compliance with the PRA.
Since FFDCA requires EPA to use validated test methods, as explained in more
detail in the procedures document, until the validation process for an assay is complete,
EPA will not be able to require any order recipient to perform that assay. Validation is
defined as the process by which the reliability and relevance of test methods are
evaluated for a specific use. The validation process was established based on
recommendations from the advisory committees consulted on EDSP implementation
over the years, and, along with the peer review materials and the validation status of
each assay, details are available on the Agency’s EDSP Web site at:
http://www.epa.gov/scipoly/oscpendo/pubs/assayvalidation/index.htm.
Completion of the validation process for all of the Tier 1 assays in the final Tier 1
battery is not a prerequisite for compliance with the PRA. Under the PRA, the Agency
is required to provide, among other things, a functional description of the information
that it anticipates will be collected (e.g., indicating how, by whom, and for what purpose
the information is to be used).2 In implementing guidance, the Office of Management
and Budget (OMB) instructs agencies to ensure that the description provided in the ICR
is sufficiently complete so as to inform potential respondents of the possible activities
they may need to engage in and any related information to be collected. For example,
1

Page 12 of OMB IQA guidance available online at http://www.whitehouse.gov/omb/inforeg/iqg
comments.pdf, states that every ICR must meet IQA guidelines. EPA’s IQA guidelines are available
online at http://www.epa.gov/quality/informationguidelines/documents/epa_infoqualityguidelines.pdf.
2
44 U.S.C. 3506(c)(1)(A)(ii); 5 CFR 1320.8(a)(2).
Page 4 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

where the specific data elements of a particular collection are expected to vary across
respondents or are otherwise uncertain at the time that the ICR is submitted, agencies
are instructed to provide sufficient information to identify the full scope of potential
information to be collected for consideration and comment by the respondents and the
public.
Under the PRA [5 CFR 1320.8(d)(1)(i)-(iv)], the agency is to seek public
comment to permit the agency to:
i.

ii.

iii.
iv.

Evaluate whether the proposed collection of information is necessary for the
proper performance of the functions of the agency, including whether the
information will have practical utility.
Evaluate the accuracy of the agency's estimate of the burden of the proposed
collection of information, including the validity of the methodology and
assumptions used.
Enhance the quality, utility, and clarity of the information to be collected.
Minimize the burden of the collection of information on those who are to respond,
including through the use of appropriate automated, electronic, mechanical, or
other technological collection techniques or other forms of information
technology, e.g., permitting electronic submission of responses.

These issues are also those about which the public may have personal
experience, perspectives and concerns that would help inform the agency while it
carries out its evaluation, and help permit the agency to make the certification required
by the PRA. This public comment opportunity is designed to help the agency in its
ongoing development of the new collection or evaluation of an existing collection.
An agency should provide the proposed collection of information in the form to
which it has been developed at that point; if the agency, for example, is at that point
considering alternative approaches to collect the information, a range of possible
questions, or different kinds of disclosure, the agency should be prepared to provide the
public with these alternative approaches. However, the agency does not have to have
completed the development of the collection of information at the point of the 60-day
advance notice. The interested public will have a second opportunity to submit
comments at the time the agency has refined the collection of information and is
submitting the information clearance package to OMB for review. [5 CFR
1320.5(a)(1)(iv).]
When the draft ICR was issued for public review and comment starting in
December 2007 and ending in March 2008, the Agency specifically identified all of the
assays that were being considered for inclusion in the final Tier 1 battery, and also
indicated that the battery was not yet finalized, and that the information collection would
be limited to the final battery – once it was established. In fact, the majority of the
assays under consideration for inclusion in the Tier 1 battery had indeed completed the
validation process, with the remaining few assays in the final stages of the EDSP
validation process. For each assay under consideration for the Tier 1 battery, the draft
ICR provided a description of the assay, its intended function in the context of the Tier 1
battery and in terms of the information being collected, and the estimated costs and
burden related to the assay. As such, the draft ICR provided the necessary information
Page 5 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

to ensure that the public and potential respondents could consider the details of the
proposed information collection activities and provide informed comment.
The public comment process for the draft ICR is not intended to serve as a
substitute for the public comment opportunities provided during the development of the
assays or related peer reviews, it is intended to provide an opportunity for the public to
comment on the Agency’s practical utility justification for the collection activities and its
related burden and cost estimates as presented in the ICR. Since the Tier 1 battery had
yet to be decided upon, the ICR identified all of the Tier 1 assays under consideration,
and clearly stated that the Tier 1 battery may consist of fewer assays, but would not
include additional assays not listed. The ICR explained the utility of the data generated
by each assay and its intended use in decision-making, as well as the estimated cost
and burden for each assay.
3. Available documentation demonstrates that the assays in the final Tier 1 battery
generate accurate and reliable results for the intended purpose.
As indicated in its July 2007 document addressing validation, EPA has
implemented the validation process for EDSP in several phases:
•
•
•
•

Preparing detailed review papers (DRPs) that involve a search of the relevant
scientific literature and development of a document that discusses the scientific
basis of each assay and critically evaluates candidate protocols.
Conducting pre-validation studies that demonstrate and optimize the assay, with
the end result being a standardized protocol for use in the multi-laboratory
validation phase.
Conducting validation studies in multiple laboratories. The purpose of this phase
is to demonstrate the transferability of the protocol, measure lab-to-lab variability,
and help establish final performance characteristics for the assay.
Peer reviewing the data to determine strengths and weaknesses of the assays.
Peer review is the critical evaluation of scientific and technical work products by
independent experts. Its purpose is to improve the quality, credibility, and
acceptability of regulatory decisions.

The documentation associated with all of these phases for each assay formed
the basis for the Agency’s determination that a particular assay generates accurate and
reliable results for the intended purpose identified for that assay, and that the assay
successfully completed the validation process.
To clarify that this documentation is available, the ICR has been amended to
clearly point the reader of the ICR to the validation documentation on the Agency’s
Website. The Agency is otherwise not required to repeat this information in the ICR.

2. Duplication of Data
Comment: Commenter #12 asserted that the EDSP draft policies and procedures will
cause unnecessary duplication of data collection, but indicated that EPA could reduce
unnecessary duplication by exempting companies from the EDSP if they meet either
Page 6 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

one of two alternative conditions. First, the company submits current information for a
chemical that indicate that the number of potential exposure pathways and the potential
for exposure are below the threshold for including the chemical on the tier 1 testing list.
Second, and in the alternative, the company submits information that is functionally
equivalent to all or some of the tier 1 assay information. Under this alternative, EPA
should only require tier 1 assays for that information for which there is no functionally
equivalent information. This recommendation is similar to the approach advised by the
endocrine disruptor screening and testing advisory committee.
EPA Response: It is important to first clarify the concept of “exemption” as it relates to
the EDSP. FFDCA section 408(p)(4) provides that “the Administrator may, by order,
exempt from the requirements of this section a biologic substance or other substance if
the Administrator determines that the substance is anticipated not to produce any effect
in humans similar to an effect produced by a naturally occurring estrogen.” The
Agency’s final policy and procedures document specifically addresses the Agency’s
approach regarding requests to exempt chemicals under FFDCA section 408(p)(4).
In the context of the first suggestion, companies were provided with an
opportunity to submit information about the exposure of the chemicals on the draft list,
including information regarding potential exposure pathways. The Agency considered
information and comments submitted in determining the final list. As such, see also the
Response to Comment documents prepared for the List and the Policy and Procedures
document.
With regard to allowing for the submission of “functionally equivalent” information
in lieu of some or all of the Tier 1 assay data, the Agency specifically addresses this
more clearly in the final Policy and Procedures document. Specifically, as under FIFRA,
EPA provides the recipients of FFDCA §408(p) test orders with the option of submitting
or citing existing data, along with a rationale that explains how the cited or submitted
study satisfies part or all of the Tier 1 Order. Existing data may include data that has
already been generated using the assay(s) specified in the Order, or “other scientifically
relevant information.” Other scientifically relevant information is information that informs
the determination as to whether the substance may have an effect that is similar to an
effect produced by a substance that interacts with the estrogen, androgen, and/or
thyroid hormonal systems (e.g., information that identifies substances as having the
potential to interact with the estrogen, androgen, and/or thyroid system(s); information
demonstrating whether substances have an effect on the functioning of the endocrine
system). Other scientifically relevant information may either be functionally equivalent
to information obtained from the Tier 1 assays—that is, data from assays that perform
the same function as EDSP Tier 1 assays—or may include data that provide information
on a potential consequence or effect that could be due to effects on the estrogen,
androgen or thyroid systems. Some “other scientifically relevant information” may be
sufficient to satisfy part or all the Tier 1 Order. The submission or citation of other
scientifically relevant information in lieu of the data specified in the Order is discussed in
Unit IV.F.1.b. of the revised Policies and Procedures document.
In addition, the Agency has written a paper entitled “EPA’s Approach for
Considering Other Scientifically Relevant Information (OSRI) under the Endocrine
Disruptor Screening Program.” This paper was developed by EPA to provide guidance
Page 7 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

to EPA staff and managers who will be reviewing the responses to Tier 1 Orders issued
under the EDSP, and may also be of interest to parties considering whether to submit
other scientifically relevant information to EPA. This paper provides general guidance
and is not binding on either EPA or any outside parties. Anyone may provide other
scientifically relevant information, and the Agency will assess the information for
appropriateness on a case-by-case basis, responding to the submitter in writing, and
making EPA’s determination publicly available. A copy of the approach paper has been
placed in Docket ID number EPA–HQ–OPPT–2007–1080.
Implicit in the comment is the idea that EPA should bear the responsibility for
making a determination of whether existing data are adequate for the EDSP prior to
issuing an order. However, both FIFRA and FFDCA clearly indicate that it is the
responsibility of the manufacturer and/or registrant to demonstrate that their chemical
and/or product can be used safely. Moreover, EPA believes that
manufacturers/registrants are better placed to identify data specific to their
chemical/product that addresses the chemical’s potential to interact with the endocrine
system.
Comment: Commenter #12 indicated that EPA acknowledges that minimizing
duplicative testing in the EDSP is a “complex issue... The agency recognizes that, if
EPA sends test orders under the EDSP screening program to multiple companies that
produce the same substance and then each recipient of the test order conducts the
required studies, there could be a great deal of duplicative testing.” 72 FR 70848 (Dec.
13, 2007). As far as we can tell, EPA’s proposed solution to the duplication problem is
to send everyone who makes the same product the same testing order, and hope that
they work something out among themselves to minimize duplication. EPA’s extensive
discussion of whether competitors will share what is often CBI suggests that the EDSP
violates the PRA’s mandate to eliminate unnecessary duplication. See 72 FR 7084870852 (Dec. 13, 2007).
EPA Response: Based on nearly thirty years’ experience with issuing data call-in
(DCI) notices for pesticide active ingredients, EPA believes that companies have
adequate incentives to join together to develop data jointly. Joint data development
minimizes the costs for each participant and EPA has almost never seen instances of
duplicative testing for pesticides. Since the EDSP procedures closely follow the
procedures used in the DCI process, EPA believes that there is no reason to expect
different results under the EDSP.
Finally, EPA believes that it is in the interest of both the Agency and industry that
orders be issued and responses documented so that all parties can clearly demonstrate
that the obligations imposed by FFDCA §408 have been met.
Comment: Commenter #14 indicated that the ICR does not account for the significant
likelihood that ambiguous test results will necessitate repeating one or more of the
assays in any battery for a particular chemical. If assays must be repeated, the costs of
even a single battery will increase dramatically. In addition, Commenter #15 stated that
the agency failed to account for the significant likelihood that ambiguous results will
necessitate repeating one or more screens for a particular substance. This is especially
likely given the lack of historical use of these assays in regulatory programs and the
Page 8 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

lack of guidance provided by the agency regarding interpretation of results. Although it
may be difficult to project the percentage of screens that might need to be repeated,
10% would be a very conservative estimate based on industry's preliminary experience
with these assays.
EPA Response: As already indicated publicly, EPA believes that the development of
tools for EPA staff, such as a Weight of the Evidence Approach (WOE) and Standard
Evaluation Procedures (SEPs), will help to provide consistency in Agency decisionmaking, as well as provide additional transparency to order recipients and the public.
With respect to the interpretation of the results from individual assays or other
data submitted or cited in response to an order, EPA is working on developing SEPs for
the initial screening, and intends to consider lessons learned in any early case-by-case
determinations. EPA intends to provide an opportunity for public review of the SEPs as
part of a peer review process.
Although the SEPs will not be publicly available in final form before EPA begins
issuing the orders, EPA expects the SEPs to become publicly available in final form
before any Tier 1 related decisions are announced to the public. EPA also expects the
SEPs to be available in draft form for public comment.
The WOE approach makes explicit the assumption that results of some assays,
in some taxa, at some level of severity, are intrinsically “worth” more than others and
should, therefore, carry more weight in decisions following Tier 1 screening. EPA will
develop reference document for interpretation of the results of the Tier 1 screening
battery which will be made public.
The SAP may also review such reference documents. This would provide the
opportunity for both written and oral public comment, but the exact process for
developing and vetting such documents has not yet been determined by the Agency.
EPA disagrees, however, that issuing test orders for Tier 1 screening cannot
occur until after such information is available in final form, or that the availability of such
information is necessary for order recipients to determine how they will respond to the
order. The information is not used to determine whether or not a chemical is on the
initial list, or to determine who should receive an order for that chemical. In terms of
responding to an order, an order recipient can certainly determine how they want to
respond to the order without considering such information.
In addition, Tier 2 assays are expected to be available for use before the Agency
announces any Tier 1 screening results, along with the information used for making
those determinations. EPA has explained publicly that the hazard and risk
assessments of a chemical will consider all available, scientifically relevant information.
The current availability of final SEPs and WOEs for EDSP related determinations does
not preclude the Agency from evaluating the potential interaction of a chemical with the
endocrine system.
Furthermore, although EPA is not currently able to provide definitive examples of
the specific circumstances in which a chemical would be able to go directly to Tier 2
Page 9 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

testing, an Order recipient may provide a justification for EPA to consider such a
request. In general, it may in some cases be possible for EPA to determine that a
particular chemical has the potential to interact with the endocrine system and therefore
could proceed to Tier 2 even if Tier 1 data are limited. However, if only some of the Tier
1 data are available to EPA, there may not be sufficient information for EPA to
determine that some of the Tier 2 data are not necessary. These determinations will be
made in a weight of the evidence judgment on a case-by-case basis and made publicly
available for consideration by others with the same or similar circumstances.

3. Practical Utility
Comment: Commenter #12 asserted that the practical utility of this ICR cannot be
determined until EPA has validated and published both the tier 1 battery of assays and
the tier 2 assays.
EPA Response: As indicated previously, this ICR only applies to the Tier 1 battery NOT Tier 2. In addition, the Agency has indicated that the individual assays in the final
Tier 1 battery will have all completed validation before the final Tier 1 battery is
announced and before any Orders are issued under the EDSP. In fact, at this time, all
but one of the assays in the proposed Tier 1 battery have completed the validation
process, and the last assay is in the final stages of the validation process. The ICR
discussions have been revised to ensure that this is clear.
Comment: Commenter #12 stated that EPA should also emphasize in the record for
both the ICR and the substantive EDSP policies and procedures that tier 1 assays have
no value unless properly validated, and that even validated tier 1 assays are useful only
as a trigger for tier 2 assays and have no other value. Consequently, EPA should
unambiguously state in the record that tier 1 assay tests should never be used for risk
assessment purposes except to trigger other assays which may be useful for risk
assessments.
EPA Response: EPA has revised the ICR to clarify the uses of the Tier 1 data
received, which are not limited in the way suggested by the Commenter. The primary
purpose of Tier 1 screening is to identify substances that have the potential to interact
with the estrogen, androgen, or thyroid hormone systems using a battery of assays.
Based on an weight-of-evidence evaluation of the available information, including the
Tier 1 data and other scientifically relevant information, the chemical will proceed to the
next stage of the EDSP where EPA will determine which, if any, of the Tier 2 tests are
necessary.
EPA has extensive experience in using data from multiple sources to develop
integrated assessments of hazard, modes of action / mechanisms of toxicity, and overall
potential for risk. EPA scientists will continue to use such experience, together with
insights from the validation process for Tier 1 assays, to address the potential of
chemicals to cause adverse effects as a consequence of interaction with the endocrine
system. In fact, EPA has considered the potential interaction of a chemical with the
endocrine system in making certain pesticide registration decisions. For example, EPA
considered data from prototypes of the assays included in the current EDSP Tier 1
Page 10 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

screen, along with other existing data, in preparing the risk assessments of
procymidone3 and vinclozolin4.

4. Methodology for Estimating Burden and Costs
Comment: Commenter #12 stated that the cost and burden of this ICR cannot be
determined until EPA has validated and published both the tier 1 battery of assays and
the tier 2 assays.
EPA Response: As indicated previously, this ICR only applies to the Tier 1 battery NOT Tier 2. In addition, whether an assay has completed the EDSP validation process
does not preclude the Agency from estimating the potential burden or related costs for
the assay. Since the tests that will be used in the EDSP are not yet offered by the
laboratories, market costs for these tests are not available. The Agency therefore used
estimated costs for 2 assays that were based on estimates provided by the EPA
scientist overseeing the validation effort for those 2 assays. Since EPA is funding the
assay validation effort, we believe that these estimates are reasonable surrogates for
actual market prices at this time and for the purposes of this ICR. For the other assays,
the Agency used the Cost Estimate Survey of commercial laboratories and other
information provided by industry representatives. Once these tests are available on the
market, these costs will be adjusted as appropriate.
Comment: Commenter #14 commented on the ICR’s calculation of the paperwork
burden for the data generation as 35% of the test costs. This percent-based estimate of
paperwork associated with conducting a test was initially established in consultation
with OMB in the 1980’s.5 In this collection, the agency therefore estimates the data
generation paperwork burden to be $131,090 per battery, per chemical. The agency
will need to adjust this estimate if more accurate and current assay costs figures are
collected. If the agency maintains a paperwork burden estimate of 35%, it can readily be
seen that this burden may more accurately total $305,440. When carried through the
calculations of table 9 of the draft supporting statement, the total burden of the
paperwork to generate the data is $22,297,120 rather than the $9,569,574 estimated by
the agency, bringing the total respondent burden to $33,389,800, rather than the
$20,662,254 estimated by the agency.
EPA Response: Once these tests are available on the market, the Agency’s estimated
costs will be adjusted as appropriate. EPA has revised the ICR to reiterate this.
Comment: Commenter #15 stated that, although it is not possible to accurately
estimate the degree to which the agency has underestimated costs without undertaking
a formal assessment, the agency's projected expenditures for the tests alone appear to
comprise less than one-half the actual costs that respondents would incur. Given the
probable financial impact of this new program, EPA should have conducted formal
3

To access the documents related to the procymidone decision, go to
http://www.epa.gov/pesticides/reregistration/procymidone/.
4
To access the documents related to the vinclozolin decision, go to
http://www.epa.gov/pesticides/reregistration/vinclozolin/.
5
From draft ICR supporting statement edspicr-draft-2007-12-05, page 27.
Page 11 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

assessments to collect actual cost data and should have conducted those cost
assessments not only for the candidate assays alone, as APT did previously, but also
for the various associated costs that were not and could not have been evaluated
previously without knowledge of EPA's implementation plans. Only in this way would it
be possible to determine whether it is reasonable to continue to assume that as much
as 35% of the test costs represent the paperwork burden.
EPA Response: Until the tests are readily available on the market, conducting a
survey about the tests would not be productive. Laboratories did not yet have any
experience with these tests upon which to they could have provided information.
Although not possible for this ICR, the Agency does intend to consult with the recipients
of these first Tier 1 Orders about their experiences, costs and burdens. This
consultation will be used to revise the estimates presented in this ICR for the ICR
renewal or future ICRs related to the EDSP.
Comment: Commenter #14 stated that the agency also assumes that 367 companies
will be required to provide tier 1 data on the initial list of chemicals and that all will
participate in a consortium but provides no rationale for these assumptions. If only 5%
of the affected entities (18) choose not to enter into a consortium, another $21,000,000
burden may be incurred.
EPA Response: This commenter incorrectly assumes that each company that chooses
not to enter into a consortium will generate the data on their own. The Policy and
Procedures document states that the Agency expects to only receive one submission of
data for each chemical. As indicated previously, EPA believes that companies have
adequate incentives to join together to develop data jointly because they routinely do so
now and have done so for the past 30 years. Companies have demonstrated in the
past that they are able and willing to join forces to minimize their costs and experience
other benefits from collaborating on the development of data for submission to EPA.
There is no evidence to indicate that companies are no longer able or willing to continue
such collaboration.
Comment: Commenter #10 believes that there are serious shortcomings with the
accuracy of the draft ICR for the following reasons, all of which are developed in more
details in the enclosed report (see EPA-HQ-OPPT-2007-1081-0010.2, APT REPORT):
• The number and complexity of tier 1 assays has increased since 2003, but EPA
did not account for this in the current ICR. Also, the agency has not made a final
determination on the design and procedures of any of the tier 1 assays.
• APT’s 2003 report did not include the costs associated with the tier 1 amphibian
metamorphosis assay or the tier 1 fish reproduction screening assay, and it did
not evaluate the potentially significant costs of analytical chemistry requirements
under good laboratory practices standards, which will be required under the
EDSP. The current ICR does not address these costs.
• There is no assurance that respondents will be expected to conduct less than the
entire battery of assays. In the absence of clarification from EPA, there is no
scientific rationale for assuming that less than the entire battery would be
required for each substance and, consequently, no foundation for a meaningful
ICR until such time as the agency provides such clarification.
Page 12 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

•

•

•

The agency has failed to account for the likelihood that ambiguous results will
necessitate repeating one or more screening tests for a particular substance.
This is especially likely given the lack of experience with these assays in
regulatory programs and the absence of agency guidance regarding
interpretation of results.
EPA’s outsourcing assumption does not overestimate the burden associated with
the EDSP as the agency claims. In fact, for the reasons APT explains, many
respondents will have to rely on contract labs and thus incur the associated
burden in terms of actual costs and paperwork.
EPA’s draft ICR does not address the significant issue of laboratory capacity.
Some of the screening assays are sufficiently new and specialized that only one
or a very few laboratories are able to perform them. Although it is certain that
supply will rise to meet demand, it is also certain that the cost of expanding
contract laboratory capacity will be borne by respondents.
Some of these comments were also reflected in comments #14 & 15.

EPA Response: The Agency accepts that the 2003 survey previously provided by
industry is not perfect or complete. It does, however, provide a reasonable estimate for
the burden and costs of these new tests. As indicated previously, the Agency did not
believe that a survey would provide more valuable information at this time because
these tests are not currently available on the market. At best, a survey might have
provided different estimates, but without a clear indication that could describe the basis
for those differences.
The ICR is not required to account for a respondent’s mistakes or inability to
follow the final protocol or test guidelines to conduct the assay. Nor is it required to
assess the capacity of commercial laboratories. An ICR is required to evaluate the
potential paperwork burden associated with information collection activities.
The Agency has made some adjustments to the ICR discussion and the test
costs that formed the basis for the data generation burden estimates. Assertions
regarding the potential for the total ICR burden and costs to be an over estimate have
been removed, and the Agency has added analytical chemistry costs to the estimate.
Please note, however, that these costs are not expected to be significant because none
of the chemicals on the List are new – such that they would not already have an
established methodology.

5. Estimated Test Costs
Comment: Commenter #10 estimates that the performing a tier 1 battery of tests may
cost $ 500,000 per substance. The acc estimates that performing tier 2 tests could
exceed $1,000,000 per substance. These cost estimates significantly exceed EPA’s
estimates for tier 1 assays.
EPA Response: There is no support for this estimate.
Comment: Commenter #14 indicated that the ICR does not account for the analytical
costs associated with each of the assays in the battery. These costs may add 100Page 13 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

300% of the agency estimated test costs to each assay within the battery significantly
increasing the total test burden. Industry representatives with experience in conducting
the assays report analytical chemistry costs in excess of those included in the updated
assay costs reported by APT. For example, for the in vivo uterotrophic assay, APT
estimates current assay costs, including a portion of the analytical chemistry costs
missing from the ICR estimates, to range from $38,000- $47,000. Industry experience
indicates the analytical chemistry costs alone can add $45,000 to this assay. Adding
this $45,000 to the ICR-estimated assay cost of $20,068 provides an estimated
minimum cost for this assay well above the $47,000 reported by APT. For the in vivo
hershberger and prepubertal assays, experience indicates the analytical chemistry
costs add $45,000-$90,000 to the cost of the assays, doubling or tripling the estimates
used in the ICR for these assays. For the in vitro steroidogensis, and
estrogen/androgen binding assays, analytical chemistry costs generally add $23,000 to
the assay costs. This quadruples the ICR estimates of the binding assays and triples
the costs of the steroidogenesis assay.
EPA Response: The Agency has added analytical chemistry costs to the ICR. The
estimates provided by APT, however, are not applicable to this ICR because they
represent costs associated with conducting analytical chemistry tests for the first time
for a particular substance, which includes initial costs to determine appropriate
methodologies. Specifically, the cost for analytical chemistry that was added to the ICR
is based on that used in other EPA ICRs that contain this test. It is important to note
that the chemicals on the initial list are all already well established existing pesticides,
which means there will not be a need to create any new analytical methods to identify
the substance for testing purposes. As such, the analytical costs included here
represent the application of the existing methods to identify the substance being tested.
Comment: Commenter #14 indicated that they believe the agency’s analysis grossly
underestimates the costs of conducting the assays in the tier 1 screening battery. In the
appended supporting document produced by APT, a review of only four of the assays
expected to be included in the final screening battery suggests that average current test
cost are 2.33 (range 1.45-3.13) times greater than those estimated by the agency. If
these increases are extrapolated over the cost of the full battery by simple calculation,
the cost of a single battery may easily reach $872,685, well in excess of the $374,543
estimated by the agency (please see the appended APT report for a discussion of
various underestimated figures used by the agency).
EPA Response: There is insufficient support for the estimates provided. For example,
because the final testing protocols were not used, the estimated costs are too high. As
indicated in the final reports for the assays, many of the protocols were revised to
minimize the use of animals or to streamline the procedures. These changes have the
effect of reducing the overall costs related to the assay using that final protocol.
Comment: Commenter #15 suggested that EPA recalculate the projected costs of the
tier 1 assays using a more formal method to account for unseen costs such as good
laboratory practices (GLP) chemical analysis for identity and purity, GLP analysis of test
concentrations, outside contractors, and in- house resources must be added to the cost
estimates to reflect the actual burden to registrants. Additionally, there will be a great
deal of registrant confusion during this “pilot phase” of the program, requiring outside
Page 14 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

guidance on optional test groups and endpoints, optional controls, studies needed, and
interpretation of the findings. Based on industry experience, consortia establishment
costs are generally in the $15k to $20k range, with ongoing costs of several thousand
dollars incurred during and after the data development phase. EPA should also
consider adding factors to account for inflation and the potential rise in laboratory costs
due to competition for approved laboratories.
EPA Response: To the extent that costs were related to paperwork burden, EPA has
increased its estimates for consortia formation and initial familiarization, as well as
added base costs for analytical chemistry. EPA will evaluate these estimates prior to
seeking renewal of this ICR.

6. Estimated Burden and Costs
Comment: Commenter #14 believes that there are several improvements that should
be made before reporting the results of the ICR to OMB for review. In both the Federal
Register ICR notice and it’s draft supporting statement, the agency states that the
estimated total annual burden for this ICR is to include, among others, “conducting
tests” however, the agency does not include the burden of conducting the tier 1
batteries in their analyses. The commenter believes that by using currently-available
test cost survey results and appropriately including this burden in the ICR, the actual
total burden of this information collection will readily approach $100,000,000. At this
time, the agency estimates the total burden for this collection to be just $20,662,254.
EPA Response: There is no basis for the $100 million estimate provided. The ICR
does include the paperwork burden for conducting the tests. In this case, the activities
are spread over the 3 year period of the ICR because there is no clear way of dividing
the activities by year. As a result, the total burden is divided by 3 to annualize it.
Comment: Commenter #14 suggests that the ancillary burdens such as those
associated with analytical chemistry, repeated assays and batteries, and transaction
and opportunity costs were not factored into the total burden.
EPA Response: These activities are not associated with the paperwork burden such
that they would ever be accounted for in the ICR. An ICR does not consider overall
impacts in the same way that an Economic Analysis would.

7. Small Entity Burdens
Comment: Commenter #14 suggested that the agency should better characterize the
burden and potential impacts on small entities to ensure that they are not unduly
affected and an appropriate mechanism is in place to address these concerns. EPA
dismisses the potential impact on small businesses by assuming that few small entities
will be covered under the EDSP and that those companies which do receive a test order
will likely join a consortium to generate the required data. However, the agency does
not adequately identify the potential data collection burden expected to be incurred by a
company that chooses to fulfill its testing obligations as part of a larger consortium nor
does EPA provide a comprehensive assessment of the burden that would likely be
Page 15 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

incurred by a company choosing to generate the data on its own. The burden to a small
entity, even if participating as a member of a large consortium, could be significant and
have a major impact on that company’s decision to remain in production.
EPA Response: There are no separate estimates for small businesses because the
burden for a small business is not expected to be different. There is no information
available to indicate that a small business participating in a consortium would
experience a greater burden. In fact, information available on small businesses
participating in testing consortiums formed under the TSCA program indicates that
these businesses experience less burden on average. EPA will attempt to identify
which Tier 1 Order recipients might qualify as a small business so as to consult
specifically with them about potentially disproportionate burdens that they experienced.
Upon renewal of this ICR or for the subsequent ICR, EPA will use revise the ICR to
reflect this consultation.

8. Other Topics
Comment: Conduct a formal evaluation. Several commenters (#13, 14, 15)
recommended that EPA conduct a formal evaluation of the test costs. Commenter #13
expressed concern that EPA has significantly underestimated the probable costs
associated with the initial phase of its EDSP, and that a formal evaluation was
necessary. Commenter #14 suggested that he agency should conduct a formal survey
of present-day, actual costs of conducting the validated assays so as to develop a more
accurate and refined assessment of “burden” (as it is defined under the PRA) that
reflects currently available data. Commenter #15 believes that a formal evaluation is
necessary because the agency has not yet issued formal protocols, test guidelines, or
other recommendations that would indicate an ultimate decision on final study designs
and procedures for any of the EDSP tier 1 assays, which directly influence costs.
EPA Response: EPA intends to conduct a formal evaluation of the test costs in
consultation with the recipients of the Tier 1 Orders.
Comment: Clarify the assays. Commenter #15 suggested that the agency should
also survey potential respondents to determine the level of detail at which assays will
actually be conducted. With the known variability inherent in some of the required and
suggested endpoints, respondents will likely feel compelled to include these optional
endpoints to hedge against what might ultimately be necessary to interpret the results.
The agency has yet to provide guidance on how results of the individual assays will be
interpreted and whether results from individual assays will be considered in isolation or
within the context of other screening studies. Furthermore, the assays are relatively new
and their stable performance unexplored. Hence, many respondents feel compelled to
include additional positive controls to better explain uncertain assay performance. The
increased costs associated with these full-featured protocols are a practical cautionary
approach against much greater costs, in both resources and animals, for repeating the
assays to obtain more complete and interpretable data. A good example of opting for a
more full-featured assay protocol involves the uterotrophic assay. According to the
current draft OECD test guideline, this screening study may be run at a limit dose or
Page 16 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

with an as yet to be determined number of test doses (2 or 3). In addition, it may include
optional endpoints, such as histopathology, as well as additional positive controls, such
as a lower potency estrogen like genistein, and additional dose groups to probe
antiestrogenic potential. In the absence of guidance from the agency on interpreting the
results, and lacking a history of reliable test performance, it is a practical reality that the
assays will include more dose groups than the minimum required by current protocols.
Hence, the costs will be much higher than projected in the draft ICR. The agency should
first conduct a formal survey to understand at what level of detail the assays are likely to
be conducted in order to develop a formal cost estimate survey based on more realistic
versions of the protocols than were available in 2003.
EPA Response: As already indicated publicly, EPA believes that the development of
tools for EPA staff, such as a Weight of the Evidence Approach (WOE) and Standard
Evaluation Procedures (SEPs), will help to provide consistency in Agency decisionmaking, as well as provide additional transparency to Order recipients and the public
because EPA intends to provide an opportunity for public review of the SEPs as part of
a peer review process.
EPA disagrees, however, that issuing test orders for Tier 1 screening cannot
occur until after such information is available in final form, or that the availability of such
information is necessary for Order recipients to determine how they will respond to the
order. The information is not used to determine whether or not a chemical is on the
initial list, nor is it used to determine who should receive an order for that chemical. In
terms of responding to an order, an Order recipient can certainly determine how they
want to respond to the order without considering such information.
The current availability of final SEPs and WOEs for EDSP related determinations
does not preclude the Agency from evaluating the potential interaction of a chemical
with the endocrine system. As indicated previously, EPA has extensive experience in
using data from multiple sources to develop integrated assessments of hazard, modes
of action / mechanisms of toxicity, and overall potential for risk. EPA scientists will
continue to use such experience, together with insights from the validation process for
Tier 1 assays, to address the potential of chemicals to cause adverse effects as a
consequence of interaction with the endocrine system.
Comment: Resolve Data Compensation & Protection Uncertainties. Commenter
#15 requested that EPA identify how and where submitted data will be maintained and
confirm if a recent suggestion that the special review and re-registration division would
be the repository of the data is correct. Another pressing question is whether the “data”
required includes inerts. Data compensation under FIFRA would be the most preferable
option, since a proven method already is in place. The FIFRA process should be
replicated for inerts that are not subject to FIFRA. EPA also needs to ensure a level
playing field at all times and establish a fair and equitable data compensation scheme
for both active and inert ingredients. One set of issues involves handling compensation
if a registrant does not support an ingredient and a group of formulators step up to take
over the testing to ensure the ingredient remains available for their product lines.
Additionally, more clarity should be provided to registrants with monitoring use products
(MUPs). EPA may choose to proceed with its proposed data compensation scheme in
Page 17 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

its initial screening. EPA must, however, revisit this issue in future implementation
notices and should gather more public and stakeholder input.
EPA Response: These issues are addressed in the Policy and Procedures document,
and are not established in the ICR.
Comment: Animal Welfare. Commenter #12 asserted that the EDSP draft policies
and procedures will cause unnecessary and useless animal death and suffering, in
violation of the regulatory acceptance criteria established by the Interagency
Coordinating Committee for the Validation of Analytical Methods (“ICCVAM”). The
commenter further “sharply criticizes EPA for allowing environmental groups like NRDC
to coerce the agency into killing many animals during tests with no demonstrated value.”
The EDSP assays must meet ICCVAM’s validation and regulatory acceptance criteria.6
These criteria require “adequate animal welfare considerations (3rs). 7 In ICCVAM’s
own words, the EDSP assays “must provide adequate consideration for the reduction,
refinement, and replacement of animal use.”8 We find no place in the record where
EPA explains specifically how it is working in the EDSP “to reduce animal use, refine
procedures involving animals to make them less stressful, and replace animals where
scientifically appropriate.”
EPA Response: This issue is related to the selection of the assays and the final Tier 1
battery. As explained in the ICR, the selection of the final Tier 1 battery is happening in
a separate but parallel process - a process that involved extensive public participation
from the very beginning in 1996. These issues are therefore more appropriately
addressed in that process, and not as part of the ICR, whose focus is solely on the
paperwork activities related to the issuance of Tier 1 Orders for the 67 chemicals
identified.
Comment: Stop allowing endocrine disruptors in the market. Commenter #04
asserted that EPA has been very negligent in allowing endless endocrine disruptors to
be bought, sold and produced in this country, when they should have been denied any
approval at all. This agency is so under the thumb of chemical profiteers that it is
disgusting. They meet constantly and this agency forgets that its biggest stakeholders
are american citizens, who provide the tax dollars for this agency to even be in
existence. All tests should be twenty years long. All tests should be on people. All tests
should show all combinant testing with every other chemical in use in the country. EPA
has allowed endless lax administration and has not been protecting the people, and life
in this country for the last 30 years. That is disgusting.
EPA Response: Although this comment is not related to the paperwork activities or
estimated burden in the ICR, EPA appreciates the concern expressed by this comment
but strongly disagrees that it has been lacking in meeting its statutory mandate to
6

Validation And Regulatory Acceptance of Toxicological Test Methods, Interagency Coordinating
Committee on the Validation of Alternative Methods, Executive Summary (March 1997). Available online
at http://iccvam.niehs.nih.gov/docs/about_docs/validate.pdf.
7
Who’s Who in the Validation of Assays for the EDSP, Briefing for New EDMVAC Members, March 2,
2005. Available online at
http://www.epa.gov/oscpmont/oscpendo/pubs/edmvac/validation_briefing_edmvac_030205.pdf.
8
Id., section 3.5.
Page 18 of 19

*** Deliberative DRAFT Document – Do Not Cite, Quote or Release ***

protect public health and the environment from unreasonable risks from exposure to
pesticides.
Comment: Consider sensitive sub-populations. Commenter #05 expressed
concern about those individuals who may have allergic reactions from exposure to
pesticides. In particular, to pesticide treatments in the office, where they felt that
residual residues existed for 2-3 days after treatment that caused them severe
reactions: including headaches, becoming lethargic, experience breathing difficulties
and sometimes develop skin irritations including hives. The following list of common
names of pesticides have been reported by scientists to be sensitizers in certain
susceptible individuals: allidochlor, anilazine, antu, barban, benomyl, captafol, captan,
dazomet, dichloropropane, dichloropropene, lindane, maneb, nitrofen, propachlor,
pyrethrum/pyrethroids, rotenone, thiram, zineb. Pesticides may be encountered as
residues in food, air and water. People may also be exposed to pesticides used in
agriculture, applications for pest control at home or at work, applications to roadside
right-of-ways to control weeds and applications of pesticides for public health vector
control programs.
EPA Response: Although this comment is not related to the paperwork activities or
estimated burden in the ICR, it is important to point out that the Agency does consider
potential disproportionate risks to sensitive sub-populations in the risk assessments that
form the basis of EPA’s pesticide registration decisions.

Page 19 of 19