September 11, 2008

Minutes of the 

United States Environmental Protection Agency (EPA)

Human Studies Review Board (HSRB)

June 24-25, 2008 Public Meeting

Docket Number:  EPA-HQ-ORD-2008-0355

HSRB Web Site:  http://www.epa.gov/osa/hsrb/

Committee Members:	(See HSRB Members list – Attachment A) 

Dates and Times:  	Tuesday, June 24, 2008, 8:30 AM – 5:15 PM

			Wednesday, June 25, 2008, 8:30 AM – 12:35 PM

			(See Federal Register Notice – Attachment B)

Location: 	EPA, One Potomac Yard (South Bldg.), 2777 S. Crystal Drive,
Arlington, VA  22202

Purpose:	The EPA Human Studies Review Board (HSRB or Board) provides
advice, information, and recommendations on issues related to the
scientific and ethical aspects of human subjects research. 

Attendees: 	Chair:			Celia B. Fisher, Ph.D. 

Vice Chair:		William S. Brimijoin, Ph.D. (via telephone)

Board Members: 	Alicia Carriquiry, Ph.D.

			Gary L. Chadwick, PharmD, MPH, CIP 

Janice Chambers, Ph.D., D.A.B.T. 

Richard Fenske, Ph.D., MPH (via telephone)

Susan S. Fish, PharmD, MPH

Suzanne C. Fitzpatrick, Ph.D., D.A.B.T. 

Dallas E. Johnson, Ph.D.

Kannan Krishnan, Ph.D.

Michael D. Lebowitz, Ph.D., FCCP

Jerry A. Menikoff, M.D.

Rebecca Parkin, Ph.D., MPH

Sean Philpott, Ph.D., 

Ernest D. Prentice, Ph.D.

Linda J. Young, Ph.D.

Meeting Summary:	Meeting discussions generally followed the issues and
general timing as presented in the meeting agenda (Attachment C), unless
noted otherwise in these minutes. 

Introduction and Identification of Board Members

	Dr. Celia Fisher (Chair, HSRB) opened the meeting and welcomed Board
members, U.S. Environmental Protection Agency (EPA or Agency) staff,
and members of the public to the June 2008 HSRB meeting.  She
acknowledged the efforts of Dr. Paul Lewis (Designated Federal Officer
[DFO], HSRB, Office of the Science Advisor [OSA], EPA) and members of
EPA’s Office of Pesticide Programs (OPP) in planning and preparing for
this meeting.  She welcomed a new Board member, Dr. Ernest Prentice,
Associate Vice Chancellor for Academic Affairs at the University of
Nebraska Medical Center.

Welcoming Remarks 

	Dr. Elizabeth Lee Hofmann (Deputy Director, OSA, EPA) welcomed Board
members, EPA staff, and the public to the meeting.  Dr. Hofmann
informed the Board that Dr. Pai-Yei Whung has recently joined EPA as
the Agency’s Chief Scientist in OSA.  Dr. Hofmann conveyed EPA’s
appreciation for the Board members’ work in preparing for and
participating in the HSRB meetings on behalf of Drs. Whung and George
Gray (EPA Science Advisor, EPA), who were unable to attend the meeting.
 She also thanked EPA staff for their efforts in preparing for this
meeting.  

	Dr. Hofmann described the topics for this meeting.  The Board will
review a protocol from the Agricultural Handlers Exposure Task Force
(AHETF) that will measure worker exposure to pesticides in a closed-cab
airblast scenario.  This review represents an important milestone for
the Board.  For the past 2 years, the HSRB has been reviewing the
outline of this proposed research.  The Board has reviewed proposed
studies, scenario designs, and field study protocols for one mop and two
wipe scenarios for the Antimicrobial Exposure Assessment Task Force
(AEATF).  Today, the Board will review research proposed by the AHETF to
measure exposure to those who apply pesticides to orchards and trellis
crops using airblast equipment drawn by vehicles with closed cabs.  The
Agency has provided governing documents and selected standard operating
procedures (SOPs) from AHETF to assist with Board review of
the protocol.  

	The Board also will review a completed insect repellent efficacy study
from ICR, Inc. (ICR).  The study evaluated the laboratory efficacy
against mosquitoes of two registered products containing picaridin.  At
the October 2007 meeting, the Board favorably reviewed this
proposed research.

	The Board will provide final review and approval of their report from
the April 2008 meeting.  EPA appreciates the Board’s efforts to make
the draft of the report available for this meeting in such a short
period of time.

Opening Remarks 

	Dr. Debbie Edwards (Director, OPP, EPA) welcomed Board members and
acknowledged Dr. Prentice as a new member of the Board.  She thanked the
Board members for their efforts in preparing for HSRB meetings and for
their thorough and thoughtful discussions of protocols and reports. 
These meetings demonstrate that human research undergoes rigorous review
both before and after it is conducted.  Dr. Edwards thanked Dr. Lewis
and his colleagues for managing the meetings and also thanked the public
for its participation.

	The Board will review a completed laboratory repellant efficacy study
with Culex mosquitoes conducted by ICR.  This is the first completed
study report submitted by this organization for Board review.  It
reports execution of a protocol the Board reviewed favorably at a
previous meeting.  EPA’s review has found the research to be
scientifically sound, but some questions related to the ethical conduct
of recruitment and slight changes to the protocol arose.

	The Board also will review a protocol from the AHETF describing field
research to measure worker exposure incurred when applying pesticides to
orchard crops using an airblast applicator drawn by a vehicle with a
closed cab.  EPA requires this data to improve its estimates of handler
exposure and risk.  EPA has found the protocol to be acceptable, but
believes that some changes, which the AHETF has agreed to make, would
improve it.

	The sampling design for these protocols has been discussed at several
Board meetings.  EPA considers the purposive diversity sampling (PDS)
strategy proposed by AHETF as acceptable if the sampling strategy is
fully documented, if random elements are incorporated whenever feasible,
and an explanation is given when incorporation of random elements is not
feasible.  For this protocol, EPA believes that more random elements
could have been included, and the lack of feasibility for incorporation
of random elements at each level of the sampling strategy was not
adequately justified; cost estimates alone are insufficient; however,
EPA has deemed the proposed strategy acceptable for this protocol, but
will expect a more thorough justification of sampling strategies in the
future.

	EPA’s review of the protocol included consultation with outside
statistical experts.  In spring 2008, EPA asked Dr. Louise Ryan (Harvard
School of Public Health) and Mr. Warren Strauss (Battelle Corporation)
to review the AHETF protocols.  The outcome of this review has been
provided to the Board.  Dr. Ryan and Mr. Strauss suggested replacement
of the proposed sampling design with a probability-based random design. 
They suggested focusing on a larger geographical area and that the Task
Force select states based on the amount of each crop type under
cultivation.

	EPA has declined to accept this recommendation because of the
significant increase in costs associated with the probability-based
approach.  In addition, even if revised there will be significant
limitations on the statistical inferences that can be drawn from the
data because of logistical issues related to creating the sampling
frame.  EPA has decided that such revisions do not sufficiently justify
the increase in costs.

	EPA will use the data from this research to update its Pesticide
Handler Exposure Database (PHED) and improve its risk assessment
activities.  The Board is asked to consider whether the proposed
research will be conducted in an ethical manner, and if the data
generated will be scientifically sound and sufficient for EPA’s
intended use.  EPA has approved the protocols and wishes this research
to go forward.

Meeting Administrative Procedures

	Dr. Lewis welcomed Board members and thanked them and his EPA
colleagues for their efforts in preparing for this meeting and also
welcomed members of the public.  As the Designated Federal Official
(DFO), Dr. Lewis serves as liaison between the HSRB and EPA and ensures
that Federal Advisory Committee Act (FACA) requirements are met.  As
DFO, he works with the appropriate officials to ensure that all
applicable ethics regulations are satisfied.  Each Board member has
filed a standard government financial disclosure form that has been
reviewed by Dr. Lewis and the OSA Deputy Ethics Officer in consultation
with EPA’s Office of General Counsel to ensure that all ethics
disclosure requirements have been met.  Dr. Lewis reminded participants
that meeting times would be approximate and that public comments would
be limited to 5 minutes.  

EPA Follow-up on Pesticide-Specific HSRB Recommendations

	Mr. William Jordan (OPP, EPA) reviewed the Agency’s actions since the
last HSRB meeting.  Topics considered at the April 2008 meeting included
the AEATF mop and wipe scenarios and protocols; ICR Protocol A382, a
laboratory study of the efficacy of an insect repellent against stable
flies; and completed studies SCI-001.4 and SCI-001.5 from Carroll-Loye
Biological Research, which tested the efficacy of two repellents against
mosquitoes.

	The Board suggested several changes to the scientific design of the
proposed AEATF mop and wipe scenarios related to the sampling design and
the duration of study tasks.  The Board agreed with the Agency that if
the proposed mop and wipe scenario design, protocol, and supporting
documentation are revised as suggested in EPA’s review, the research
would meet the applicable requirements of 40 Code of Federal Regulations
(CFR) part 26, subparts K and L.  In response to HSRB recommendations,
EPA and the AEATF are developing revisions to the mop and wipe protocols
to address EPA and HSRB comments and the AEATF expects to execute the
research later in 2008.

	The Board’s concerns and recommendations for ICR Protocol A382 were
mainly related to ICR’s lack of recruitment of non-White subjects.  If
the protocol is amended consistent with the Board’s concerns and
recommendations, particularly regarding recruitment of under-represented
minorities, this protocol would be sufficiently sound, from a scientific
perspective, to be used to assess the repellent efficacy of two
formulations of picaridin against stable flies.  The HSRB agreed with
the initial assessment of the Agency that, if the protocol is revised as
suggested by EPA and the HSRB, the study submitted for review would meet
the applicable requirements of 40 CFR part 26, subparts K and L.  In
response to HSRB recommendations, ICR has consulted with EPA and revised
the protocol and consent form to address EPA and HSRB comments.  The
revised proposal will be submitted in July 2008 for Institutional Review
Board (IRB) review and ICR expects to execute the study in August 2008,
and submit the completed study report for HSRB review in January 2009.

	The HSRB agreed with the Agency that the completed studies SCI-001.4
and SCI-001.5 from Carroll-Loye Biological Research were sufficiently
sound, from a scientific perspective, to be used to accurately calculate
the complete protection time of the test materials for repelling
mosquitoes.  The HSRB concurred with the initial assessment of the
Agency that the study submitted for review by the Board meets the
applicable requirements of 40 CFR part 26, subparts K and L; EPA will
rely on data from studies SCI-001.4 and SCI-001.5.

	The Board will review additional completed studies from Carroll-Loye
Biological Research in the future.  The HSRB reviewed the protocol for
the laboratory tick repellent study SPC-002 in October 2007 and the data
collection for this study was completed in March 2008.  The HSRB
reviewed the protocols for the field mosquito repellent studies SPC-001
and LNX-001 in October 2007 and June 2007, respectively, and data
collection for these studies was completed in early June 2008.  All
three completed studies are expected to be reviewed by the HSRB in
October 2008.

AHETF Pesticide Handler Protocols: Closed-Cab Airblast Scenario

Background, Scenario Definition, and Sample Selection

	Mr. John Carley (OPP, EPA) presented background information on the
AHETF closed-cab airblast scenario protocols.  These represent two of
five planned field study protocols for measurement of potential dermal
and inhalation exposure during application of liquid pesticides to crops
using conventional airblast sprayers drawn by vehicles with enclosed
cabs.  The AHETF submitted an IRB-approved scenario design of these two
field study protocols for this scenario on April 7, 2008, along with
extensive supporting documentation.  EPA’s Science and Ethics Reviews
of May 27, 2008 were based on review of the April 2008 submission. 
The AHETF responded to EPA’s review on June 13, 2008.  The remaining
three field study protocols associated with this scenario will be
considered at the HSRB’s October 2008 meeting.

	These are proposals for research involving intentional exposure of
human subjects, with the intent to submit the resulting data to EPA
under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA);
therefore, 40 CFR §26.1125 (which requires prior submission of the
protocol and supporting documentation) and 40 CFR §26.1601 (which
requires review of the protocol by EPA and the HSRB) apply.  The
AHETF’s April 7, 2008 submission included the following:

documents describing the closed-cab airblast scenario design

Florida citrus crop protocol AHE55

Georgia pecan protocol AHE56

AHETF governing documents

governing documents with tracked changes

referenced SOPs and other referenced materials

AHE55 IRB materials

AHE56 IRB materials

The AHETF governing documents and referenced SOPs were not discussed at
this meeting.

	Many elements of the two proposed protocols are the same.  Both are
field study protocols that address the same scenario and implement the
same scenario design.  The protocols are identical in substance; the
field phases differ only in location and crop.  The analytical phases of
the protocols are identical.  This is the first scenario-specific design
for agricultural handler exposure studies to be reviewed by the HSRB. 
These protocols represent the first responses from the AHETF to EPA’s
November 2007 decision to accept purposive sampling incorporating random
elements whenever feasible, with explicit conditions.  The April 7, 2008
protocol submissions contained all elements of documentation required by
40 CFR §26.1125.  Although some refinement is needed to address the
comments in EPA’s Science and Ethics Review, EPA has considered these
proposals to be ready for HSRB review.

	The closed-cab airblast scenario is defined as application of liquid
sprays by conventional airblast sprayers to actively growing, foliated
orchard trees and trellis crops, using closed-cab tractors.  It excludes
mixing and loading for application, use of unconventional airblast
sprayers (i.e., use of electrostatic technologies or other methods for
minimizing non-target dispersion), airblast application with open-cab
tractors, and application to dormant crops.  The conventional airblast
sprayer is composed of a tank carrying the pesticide mixed with water,
comprising the one-third of the sprayer closest to the tractor, and a
large fan, comprising the rear two-thirds of the sprayer.  Nozzles that
distribute an atomized spray can be found on either side of the sprayer
and when both sets of nozzles are working, the sprayer and tractor can
be enveloped by a cloud of atomized droplets.

	The closed-cab airblast application scenario is one of 33 discrete
pesticide mixing/loading or application scenarios defined by AHETF. 
This scenario includes five clusters, each of which is associated with a
separate field study to be conducted in a different part of the country
with a different crop.  The two protocols reviewed during this meeting
were the Florida citrus and Georgia pecan protocols.  This application
scenario differs from the mop and wipe studies reviewed by the HSRB at
previous meetings.  In contrast to the mop and wipe studies, in which a
single study protocol could provide all the monitoring data needed to
populate the scenario design because the research was conducted indoors,
the monitoring data for the closed-cab airblast scenario will be
collected in different parts of the country under five different field
study protocols.  Agricultural research must include the growers.  To
obtain realistic data, agricultural handler exposure studies require
that agricultural pesticides be applied to appropriate crops using
appropriate equipment, which can be accomplished only within the terms
of registration of the pesticides and with the consent of the growers to
whose crops pesticides are applied.  A critical complication in the
sampling design of these studies is the need to identify growers who are
eligible and willing to cooperate in the research before recruitment of
workers can begin.

	No more than two monitoring units (MUs), or workers/applicators, can
come from any one grower.  As many as five growers may be selected for
each cluster.  

	Initial purposive design choices include the following:

selection of crops and states by choosing from states with the highest
production of crops typically sprayed by airblast equipment

selection of crop types by height and cultural practices

exclusion of dormant application

exclusion of hops and other unusual crops

selection of no more than one crop type per state

selection of no more than one state per agricultural region.

	The first stage of diversity selection involves establishment of
clusters.  Crops for which airblast application is typical are
identified; because airblast equipment is designed for use in orchard
and on trellis crops, identifying the crops where it is likely to be
used is straightforward.  Next, the predominant states in which such
crops are grown are identified using United States Department of
Agriculture production statistics.  States are grouped by agricultural
region to incorporate diversity in conditions and agronomic practices;
this is based on “agricultural regions” as defined by EPA when food
residue or environmental data are needed from different parts of the
country where different agricultural conditions and practices exist. 
One state for each crop type and one state in each agricultural region
will be selected to maximize diversity with respect to region, climate,
agricultural practices, and equipment.  Although this approach increases
diversity, it sacrifices representativeness.  For example, although
California has more than half the acreage in the United States in three
of the identified crop types (nuts, stone fruits, and trellis crops), it
is sampled only once.  Next, a specific monitoring site (county or
counties) within each selected state is chosen to minimize the costs of
gathering needed monitoring data.

	To conduct the study efficiently, a specific area within each selected
state where growers and workers can be recruited is identified. 
Limiting the geographic area in which the protocols will be executed
allows grower and MU identification and selection to be conducted
efficiently from one local base.  Collecting all MUs during a single
visit to the study area minimizes researcher salaries, travel, food,
lodging, and field fortification expenses.  In its June 13, 2008
response to EPA, the AHETF argued that although two MUs can be collected
from workers associated with a single grower in a single day, MUs from
two different growers could not be collected on the same day.

	The primary considerations for selection of study counties will be the
availability of the target crop sprayed with airblast equipment; a large
pool from which to recruit suitable growers who would be willing to use
the AHETF surrogate compounds and participate in the study; and a Local
Site Coordinator (LSC) with experience conducting similar studies and
who is familiar with local agricultural practices.  The likelihood that
growers will agree to participate in an AHETF field study is influenced
by how and by whom they are approached.  Thus, a LSC who is familiar
with both the research and the local growers will facilitate
recruitment.  EPA agrees that a locally knowledgeable LSC is important,
but believes strongly that the LSC should not also be a participant in
the research, either as a grower or employer of workers who will be
monitored.

	The second stage of diversity selection identifies MUs.  This involves
1) defining practical strata for the amount of active ingredient handled
(AaiH); 2) identification and recruitment of a pool of eligible growers
with appropriate workers and equipment; 3) design of an array of MUs
while considering the size of the equipment and area to be treated,
AaiH, and other parameters; 4) approach and recruitment of eligible and
interested workers to complement the MU design; and 5) repeating these
steps as needed to complete the design.

	Stratification by AaiH is achieved by defining five bands, in range
from 5 pounds to 100 pounds AaiH per day.  A minimum of 5 pounds AaiH
per day will reduce non-detects to a minimum and 100 pounds AaiH per day
is the practical maximum for a typical work day.  In addition, EPA
believes that it can extrapolate upwards from 100 pounds AaiH accurately
enough for its risk assessment activities.  One MU for each AaiH stratum
will be identified.

	At this point, all choices in the process (crop type, state/region,
site/county, AaiH strata, etc.) have been purposive, but selection of
cooperating growers will employ a random process.  First, eligible
growers will be identified by developing lists of local growers of the
crop of interest based on information from knowledgeable local
resources.  Local resources are defined as LSCs, commercial applicator
firms, university/county extension agents, crop consultants (e.g., pest
control advisors or commercial applicators), agricultural researchers
including LSCs, chemical dealers or sales representatives, and grower
associations.  All lists will be combined, duplicate entries suppressed,
and the list randomized.  Growers will be contacted in the order they
appear on the randomized list to determine interest and apply
eligibility factors.  Willing and eligible growers will be placed in a
“working pool” until sufficient numbers of growers have been
identified.  In its June 13, 2008 communication, the AHETF notes that
the list of growers may not be complete, but that it should be
“sufficiently complete for the purposes of this research.”  If the
list is significantly incomplete, the resulting sample will be
significantly non-random.  EPA has recommended that the AHETF define
SOPs for all steps within the sampling process; the AHETF has agreed to
develop these SOPs.

	Growers placed in the working pool must meet the following
requirements:

grow the target crop

have sufficient acreage to support use of the minimum AaiH in a single
day

spray the crop with conventional airblast equipment drawn by vehicles
with enclosed cabs

have at least one worker experienced in using closed-cab airblast
equipment

be willing to cooperate in the research

be willing to use at least one of the AHETF surrogate chemicals

be willing to let the AHETF recruit workers without interference or
influence.  

The last three items are related to the growers’ willingness to
participate.  Because neither the AHETF nor EPA can compel a response,
it will not be possible to assess non-response bias by learning more
about growers who are either uninterested or ineligible.

	Data compiled from growers in the working pool will include the
following:

a list of crop(s) and acreage that might be treated

specific location of crop(s)

number, type, and size of airblast equipment available

surrogate chemical(s) that might be used

approximate timing of expected treatment

number of experienced workers available

the amount of active ingredient those workers might be able to handle in
a day given equipment and acreage.  

The amount of active ingredient (AI) potentially handled will be
estimated by the investigators based on information provided by the
grower about his spray equipment and acreage of the target crop.

	The third step proposed in “2nd-stage diversity selection” is the
design of what the AHETF calls an “efficient configuration” of MUs. 
The LSC and study director will define a group of MUs (i.e., growers,
chemicals, workers, AaiH, timing) that would result in an efficient
study.  Such a configuration would include growers within a compact
geographical area that expect to apply one of the surrogate pesticides
at around the same time using airblast sprayers of differing size and
type and who could together provide workers for all defined strata of
AaiH.  An efficient configuration must include more workers than the MUs
required to satisfy the study design.  EPA has questioned how many more
workers than MUs are needed.  If there were three or four qualified
workers for each MU and if more than one of them were willing to
volunteer, the individual worker to be monitored could be selected
randomly.  The AHETF agreed to define an efficient configuration with at
least 3 or 4 potential workers for each defined MU, and to randomly
select the workers to be monitored from among qualified volunteers.

	To approach and recruit qualified workers, growers will be visited to
confirm their eligibility and willingness to cooperate and to obtain a
promise of non-coercion.  Qualified workers employed by eligible,
cooperating growers will be informed about the research and asked if
they wish to participate.  These actions will continue until the MU
design is filled, with the following constraints:  no worker is used
for more than one MU; no airblast sprayer is used more than once; and no
more than two MUs will come from any single grower.  When multiple
qualified and willing workers are employed by a single grower,
participants will be selected randomly.

	The third constraint (no more than two MUs from a single grower) was
not explained in the AHETF proposal submitted on April 7, 2008.  In its
review EPA recommended this be changed to “no more than 1 MU from any
one grower.”  In its response of June 13, 2008, the AHETF argued that
to change from three to five growers would significantly increase the
logistical effort and cost of the field study and is not necessary to
ensure sufficient diversity of MUs to meet the goals of the research. 
The AHETF proposal lacks details about how the third constraint would be
applied if workers are employed by a commercial spray service.  Even if
a service that works with several growers and has five qualified and
experienced workers and five different airblast sprayers, all five MUs
should not be drawn from the same employer.

	The sampling design can be summarized as:  1) choose study crops, study
areas, and AaiH strata; 2) identify growers of the target crop in the
study area; 3) approach growers in random sequence; 4) compile data from
eligible growers willing to cooperate; 5) design an efficient
configuration of MUs; and 6) recruit a worker for each MU, selecting
them randomly when multiple qualified workers are employed by the same
grower.  In its June 13, 2008 response to EPA, the AHETF has agreed to
over-recruit workers for each MU to permit random selection of the
worker to be monitored; however, the AHETF’s goal of minimizing the
number of growers involved may compromise the randomness of worker
selection.  This part of the process needs to be specified in an
appropriate SOP.

	The AHETF has responded to EPA requests in its November 2007 decision
to accept PDS for both the AEATF and AHETF exposure monitoring programs
to describe in detail their sampling design for each scenario and
incorporate random elements whenever it is feasible.  The AHETF
described in detail the sampling design for the closed-cab airblast
scenario in its April 7, 2008 submission.  The AHETF also agreed to
over-recruit growers and workers to allow random selection of monitored
workers.  The AHETF was asked to document its rationale for using a
particular approach, including all decisions regarding the feasibility
of randomizing specific elements in the design.  The AHETF has provided
some documentation of its rationale, but provided only negligible cost
data to document the feasibility or infeasibility of alternative
designs.  

	In its first submission in response to EPA’s November 2007 guidance,
the AHETF has provided enough information to permit the Agency to
conclude that the sample selection process proposed for this scenario
design and these protocols is acceptable, with minor refinements.  EPA
expects to receive a more complete rationale for the sampling choices
made in future AHETF scenario designs and protocols.  The AHETF promised
in its letter of June 13, 2008 both to make the refinements EPA
requested from its review and to provide a more complete rationale for
the sampling design and sample selection process in future submissions. 
EPA believes the proposed sample selection process, as altered
consistent with the AHETF’s June 13, 2008 response to EPA’s review,
is acceptable.

	Clarifying Questions

	Dr. Richard Fenske asked for clarification whether the charge to the
Board asks it to assess exposure to handlers who apply liquid
pesticides, but Mr. Carley’s presentation indicates exposure to liquid
sprays will be assessed.  Mr. Carley clarified that liquid spray was the
correct term.  At least one of the seven products to be tested is a dry
formulation in water soluble packets, but all products will be mixed
with water to produce a spray.  Because the scenario does not include
exposure occurring during the mixing and handling processes, the
original formulations of the pesticides are not relevant.

	Dr. Michael Lebowitz said that he was impressed by the use of the
existing databases to design the studies, but noted that in his own
perusal of the data he did not find any evidence indicating a
relationship that the AHETF could have used to determine the amount of
AI needed to minimize non-detects in these applications.  This is
critical for the AaiH strata scheme, which is unevenly constructed.  Dr.
Lebowitz also expressed skepticism at EPA’s claim to be able to
extrapolate beyond 100 pounds of AaiH.  He asked if EPA knew the levels
of AaiH that produced non-detects.  Mr. Carley answered that the AHETF
would need to answer this question.  Dr. Fisher inquired about the
relationship between AaiH and pounds of AI, given the water-soluble
nature of the products.  Mr. Carley explained that regardless of initial
physical form and concentration, the amount of AI sprayed can be
determined based on the amount of product placed in the tanks and by
calibration of the equipment and speed of the sprayer.  The AaiH strata
are based on the number of pounds of AI handled in a given monitoring
event.  Dr. Alicia Carriquiry asked if stratification based on AaiH was
based on what workers are anticipated to handle.  Mr. Carley answered
that the stratification is by range of AaiH, not precise amounts.  The
AHETF designed these strata to distribute the data points and improve
the efficiency of the design.  The amount of crop acreages, type of
sprayer, volume of sprayer, and speed of sprayer through the crop are
used to select growers.  For example, a small grower expected to take
three days to apply pesticide to his crop likely will fulfill the lowest
band of AaiH.  Because the tanks of the sprayers will be emptied, the
amount of AI sprayed can be determined.

	Dr. Lebowitz noted a lack of sufficient use of agricultural extension
agents and their significant knowledge about local growers, types of
crops, and characteristics of growers; extension agents also work
independently of growers.  Agricultural extension agents could help
identify potential participating growers.  He also asked about the lack
of clarity in the SOP regarding exposure of the AI to the MU at the time
of mixing and loading and if MUs will be near enough to the mixing and
loading site to be exposed to the pesticide during these processes.  Mr.
Carley answered that the AHETF representatives would describe the
location of the MUs during mixing and loading operations.  He also
explained that agricultural extension agents are on the list of local
experts that the AHETF expects to consult.  Dr. Lebowitz added that the
agents could have information that would help determine why certain
growers decline to participate.

	Dr. Janice Chambers inquired whether the lowest AaiH stratum was
expected to produce detectable residues.  Mr. Carley responded that at
the lowest threshold of 5 pounds AI, there likely will be detectable
residues.  Dr. Chambers asked if smaller growers would qualify for only
the lower AaiH strata.  Mr. Carley indicated that this was not clear. 
There is a relationship between AaiH strata and the capacity of the
sprayer and size of the orchard that needs to be clarified by the AHETF.
 Dr. Chambers queried whether the AHETF believed it would be able to
convince growers to use one of the two surrogate pesticides designated
for use in the research protocols.  Mr. Carley explained that the two
surrogate compounds are registered and widely used.  If a grower agrees
to use one of these products, they can pick the product and physical
form that best suits their equipment.  These pesticides were chosen
because there are robust analytical methods available for them and they
are relatively non-toxic, requiring use of less personal protective
equipment (PPE).  Dr. Chambers inquired if participating growers would
be required to spray their crops regardless of need.  Mr. Carley replied
that the AHETF intends to select growers who anticipate applying
pesticide within a certain timeframe.  Dr. Chambers questioned why the
AHETF is asking the growers to buy the pesticide and be reimbursed,
rather than having the Task Force purchase and supply the pesticide. 
Mr. Carley remarked that members of the Task Force would need to answer
that question.

	Dr. Dallas Johnson inquired about options if, as growers are
randomized, there are insufficient workers among the three growers to
obtain sufficient numbers of MUs.  He asked if it would be possible to
stratify based on the number of workers employed by each grower. 
Mr. Carley explained that to do this, the phase of the design and
efficiency of the configuration must be considered.  He added that it
would not necessarily violate the random order of selection.  Dr.
Johnson asked about the impact of the first five growers on the random
list having only one worker.  Mr. Carley responded that there will be
more than three growers in the working pool, and emphasized EPA’s
recommendation that a SOP was needed to clarify this stage of the
sampling process.

	Dr. Fisher summarized the Board’s questions for the AHETF.  The Board
asked to be informed about data that assures that lower levels of AaiH
strata would have detectable residues, whether the workers would be
exposed during the mixing and loading process, whether small growers
would be restricted to the lower AaiH strata, whether pesticide would be
applied even if it was not needed, and why growers will be asked to
purchase the pesticides and then be reimbursed.

Sample Size, Sampling Design, and Cluster Configuration

	Mr. Matthew Crowley (OPP, EPA) provided EPA’s review of the
closed-cab airblast scenario sampling size, sampling design, and cluster
configuration; EPA’s regulatory objective for this study is to obtain
data that will enable the Agency to predict exposure for closed-cab
airblast application of any pesticide AI.  By estimating AaiH using
application rates and acres treated, EPA believes it will be able to
reasonably predict exposure to assess closed-cab airblast applicator
risks.  AHETF’s primary objective is to provide a distribution of
normalized unit exposures (e.g., milligrams (mg) AI per unit of AI
handled) for workers applying pesticides with closed-cab airblast
equipment.  AHETF’s secondary objective is to distinguish between
complete proportionality and complete independence between closed-cab
airblast application exposure and AaiH.  If the analysis shows that use
of the unit of exposure in the form of mg per AaiH is unreasonable for
the purposes of risk assessment, an alternate normalization factor (or
none at all) can be examined.  Determining the extent of a particular
factor’s influence on exposure (such as crop type, wind speed, number
of entrances or exits from the cab) would require different sampling
designs and will not be assessed in this research.

	The sampling design proposed by the AHETF is referred to as diversity
selection.  Its purpose is to obtain, given a small sample size, a
diversity of conditions expected to directly or indirectly influence
exposure.  A MU is equivalent to a handler-day and involves one worker,
one closed-cab airblast sprayer, one AaiH stratum, and one day’s work
(or application of at least three tank-loads of pesticide).

	In its consideration of the sampling design, the EPA consultants Dr.
Taps Maiti (Iowa State University), Dr. Ryan, and Mr. Strauss have all
emphasized the relative desirability of a probability-based sampling
design.  Dr. Ryan and Mr. Strauss suggested an alternative using a
probability-based random selection of the study areas and monitored
workers.  This design attempts to incorporate probability-based sampling
for a sample of equivalent size to that proposed by the AHETF, without
substantially altering the operational efficiency of the research; the
alternative approach to selecting areas for monitoring calls for
selecting two states per crop type to ensure diversity of geographic
location.  States and counties would be sampled with a probability
proportional to the number of bearing acres.  The approach mandates that
a state with more than 50 percent of the bearing acres would always be
selected as the 3-MU state, a state with 30 to 49 percent of the
bearing acres would always be selected as the 2-MU state, and a state
with less than 30 percent of the bearing acres would be selected with a
probability proportional to its share of bearing acres.  Once states are
selected, counties within these states would be chosen with a
probability equal to the proportion of bearing acres of the target crop
in each county of the state.  Using this approach, Florida would have
three MUs for citrus crops and California would have two.  California
would have three MUs for nut crops and Texas would have two.  This
approach results in over-representation of California.

	After states and counties are selected, eligible within-county growers
of the appropriate crop will be contacted to gather information about
likely closed-cab airblast applications and AaiH within the target date
range.  Eligibility will be based on objective grower characteristics
(e.g., sufficient acreage, at least one experienced closed-cab airblast
application worker) rather than their willingness to participate. 
Unwillingness to participate should be treated as part of non-response
and analyzed as such.  Assuming all (or a high proportion of) eligible
members of the sample provide information about their pesticide use, the
highest stage sampling frame will be stratified (workers within
geographic areas selected within the lower stages of the design) and a
sample selected.

	EPA’s concerns regarding this proposed design and for
probability-based sampling designs for agricultural pesticide handler
monitoring in general are related to sampling frame, selection of field
study locations, selection of growers and workers, and non-response
bias.  The ideal sampling frame would be based on a census of
acre-treatments by closed-cab airblast equipment; unfortunately, this
information is unavailable.  Possible surrogates from which to generate
a sampling frame include bearing acreage (e.g., 50 percent of nut tree
acreage in the United States are in California); treated acres (e.g., 50
percent of nut tree acreage in California is treated at least once per
season with pesticides); or acre-treatments (e.g., 50 percent of
pesticide-treated nut tree acres in California are treated three times,
20 percent are treated twice, etc).  Information on bearing acreage is
easily available, but incomplete as not all bearing acres are treated,
not all are treated using airblast sprayers, not all airblast sprayer
treatments are delivered using closed-cab vehicles, and not all acres
are treated the same number of times.  Treated acres can usually be
estimated with good confidence, but not all treated acres are treated
the same number of times or with closed-cab airblast equipment.  Basing
the sampling frame on acre treatments is most nearly ideal, but would
rely more on expert judgment.  Many factors (including weather, pest
pressure, equipment availability, etc.) can affect whether, how often,
and with what pesticide a given acre is treated.  Constructing an
appropriate surrogate sampling frame is possible, but the more the
sampling frame approaches the ideal, the more expensive it would be and
the more it would depend on expert judgment.  Additionally, use of a
surrogate to generate a sampling frame would provide unknown inferential
power.  EPA considers the inability to construct an accurate sampling
frame to be a fatal flaw for a probability-based design.

	EPA also has concerns related to selection of field study locations. 
Several practical considerations indicate the need for purposive
selection of field study locations.  In order to contain costs, field
study locations should be in contiguous, compact geographical areas
where growers are likely to spray the target crop at approximately the
same time.  The selected counties also should have many potential
cooperating growers of the target crop to allow diversification of
equipment and other factors, and random selection of monitored workers. 
The number of field study locations is a major cost factor in this kind
of research program.  If a probability-based approach results in
selection of geographically dispersed growers, the study would be
significantly more expensive and less efficient.  Monitoring in two
states for each crop-type would potentially double the number of field
study locations for the closed-cab airblast scenario from 5 to 10, which
EPA estimates would cost about $100,000 for each additional location.

	Another issue of concern for the probability-based sampling approach is
selection of growers and workers.  Gathering information on anticipated
handler-days from growers within the selected counties as Dr. Ryan and
Mr. Strauss suggested is likely to be very difficult.  For this
research, growers and workers are not independent populations; growers
willing to cooperate must be identified before workers can be recruited
or selected.  In addition, although states maintain records of certified
pesticide applicators, there is no registry of non-certified pesticide
handlers, and neither the surrogate chemicals proposed nor many other
pesticides applied by closed-cab airblast equipment are of sufficient
toxicity to restrict their use to certified applicators.  The
information needed to develop a list of all possible or most pesticide
handlers who might make closed-cab airblast applications is not
available.

	EPA also has concerns about the impact of non-response bias.  Response
rates are typically low in pesticide worker exposure research; only a
small percentage of eligible growers are willing to provide information
about pesticide use or to cooperate in the research.  Neither EPA nor
the AHETF can compel participation, which makes it difficult (or
impossible) to assess non-response bias.  In their proposal, Dr. Ryan
and Mr. Strauss assumed that participation will be high and that the
growers will readily supply information and participate in the research.
 However, this has not historically been the case.  Dr. Ryan and Mr.
Strauss suggested that unwillingness to participate should be treated as
non-response and analyzed as such, but neither EPA nor the AHETF can
obtain information about why a grower would or would not choose to
participate.

	EPA acknowledges the theoretical benefits of probability sampling but
has concerns about the many practical barriers to a fully
probability-based sampling design for the AHETF monitoring program.  The
Agency has concluded that diversity selection incorporating random
elements whenever feasible is an acceptable compromise; however, for
these protocols, EPA believes that the AHETF failed to adequately
document the infeasibility of probability sampling.

	Factors considered by the AHETF for the proposed sample size and
cluster configuration for the closed-cab airblast scenario include a
lack of usable existing closed-cab airblast applicator MUs; a target
accuracy benchmark (or K-factor) in which selected log normal-based
estimates are accurate to within 3-fold at least 95 percent of the time,
of less than or equal to (≤) 3; diversity of crop types treated with
closed-cab airblast equipment; maximization of efficiency; and
minimization of costs.  Development of the K-factor of ≤ 3 relied on a
reference model with the same two-stage nesting structure as the
proposed sampling approach, but an assumption of simple random sampling
at each stage.  Thus, the reference model does not reflect the hybrid
combination of purposive and random diversity selection proposed, but is
nonetheless expected to be useful to help estimate an appropriate sample
size.  The reference model assumes that normalized unit exposure is
log-normally distributed based on existing pesticide handler exposure
data conducted before new EPA rules were implemented in 2006.  Analysis
of these completed AHETF studies found a geometric standard deviation
(GSD) of approximately 4 and an intra-cluster correlation (ICC) of
approximately 0.3. 

	The proposed sample size and cluster configuration for this research is
five clusters of five MUs.  This configuration achieves the target
accuracy benchmark at the 95th percentile under the assumptions that GSD
= 4 and ICC = 0.3.  This approach reflects the diversity of the
closed-cab airblast scenario; the five selected crop types represent
approximately 95 percent of all acreage potentially treated by airblast
equipment.  The structure of one crop per state and one state per region
helps to maximize efficiency and minimize costs.  The 5x5 configuration
will generate the data the AHETF needs when costs and benchmark accuracy
targets are considered, but the configuration can be changed if another
is found to be preferable.  Other configurations that could achieve
accuracy equivalent to the 5x5 configuration include 15 clusters of one
MU.  This is the smallest sample that achieves target accuracy and
minimizes the ICC; however, gathering data from 15 different clusters
would be prohibitively expensive.  A configuration of four clusters of
eight MUs would provide a larger total sample, although one less
cluster, at comparable costs.  However, this is inappropriate for the
closed-cab airblast scenario because data would not be gathered for one
of the five major crop types.  Thus, EPA has concluded that the 5x5
configuration is acceptable for the closed-cab airblast application
scenario.  The resulting data will be of improved quality and quantity
over existing data and the sample size exceeds the minimum called for in
EPA and the Organization for Economic Co-operation and Development
(OECD) Guidelines.  The costs associated with including additional
clusters could reduce AHETF’s ability to monitor additional scenarios.
 In addition, if the benchmark accuracy target is not met, additional
data collection will be considered.

	The Agency has concluded that an overall probability-based random
sampling design is technically and financially infeasible for this
scenario; therefore, diversity selection incorporating random elements
whenever feasible is an acceptable compromise.  EPA recognizes that the
resulting distribution will not be statistically representative but
believes that the resulting data will improve risk assessment; however,
EPA expects better cost analysis and justification for sampling design
in future AHETF scenario designs.

	Clarifying Questions

	Dr. Carriquiry noted that it was not clear if the five MUs would be
obtained from different growers.  Choosing three growers and picking two
workers from each of two growers could introduce an additional
correlation not taken into account in the sample size calculations.  If
two workers are chosen from each of three growers, the sample size will
need to be increased. 

	Dr. Fenske inquired about the basis for EPA’s conclusion that
response rates for participation in pesticide research are typically
low.  Mr. Crowley explained that this is a claim of the AHETF and it is
supported by the Agency.  Mr. Carley clarified that this information is
contained in the background materials for this research and was used to
inform EPA decisions regarding sampling in November 2007.  Dr. Fenske
asked why EPA believes that study investigators cannot assess reasons
for grower non-participation.  Mr. Crowley responded that if growers
decline to participate because they are breaking laws regarding
pesticide handling or other matters, EPA will not be able to learn this.
 It has not been possible to receive clear answers for reasons for
non-participation.  Dr. Fenske inquired if the investigators will ask
growers who decline why they are declining.  Mr. Carley explained that
the concern is that reasons for non-participation could be collected,
but EPA is unclear about how these answers would influence the results. 
Dr. Lebowitz noted that in population sampling, it is often difficult to
obtain sufficient information from non-participants, but investigators
can compare characteristics of responders and non-responders.

	Dr. Lebowitz commented on the issue of the different number of crops
included in the closed-cab airblast scenarios.  He said that it is
understandable that there may be differences in exposure based on crop
type, but the Board has seen no data for this.  He stated that it was
unclear why these five crop types were chosen and others were not, nor
was it clear why it was necessary to include all five of these crop
types.  He also explained that good repetition of at least the 50th to
the 99th percentile of exposures and small uncertainty factors are
needed for risk assessment.  Ensuring that data will be gathered for the
top half of the distribution is critical, but does not seem to be
addressed in the diversity sampling approach.  Mr. Crowley remarked that
the rationale for choosing five different crop types was based on the
ICC and modeling used to develop the sample size and configuration.  The
argument for the ability of diversity selection to capture the upper
half of the distribution can be found in the Task Force governing
documents.  Mr. Crowley added that he was unsure if this argument was
accurate; capturing a wide variety of factors that may or may not
influence the results and affect bias could result in over-estimation at
the upper percentiles of exposure and under-estimation at the lower
percentiles.  Mr. Carley explained that there are important differences
among the crop types; vines and trees on which these crops grow are of
different heights and are spaced differently in the orchards.  Different
equipment is used to spray the crops, which are also grown in different
parts of the country with different agronomic practices, ambient
temperatures, and humidity.  The AHETF picked the five crops for five
clusters by observing that nearly all airblast applications are made to
these five crops and they are grown in widely distributed geographical
areas.

	Dr. Prentice asked about the number of workers expected to agree to
participate.  Mr. Carley noted that this would be discussed in more
detail in his ethics review of the protocol.  Previous studies have not
reported this information, but the Task Force plans to collect this
data.

	Dr. Kannan Krishnan noted that the most important factor affecting
exposure would be the application rate.  Given the different crop types,
this is likely to vary substantially; PDS could be informative for this
aspect of the study.  If application rate also differs by type of
pesticide, the data would be improved.  He asked if the number of pounds
AaiH reflected the number of acres treated.  Mr. Carley replied that
application rates vary based on equipment type and crop type.  AaiH
versus application rate per acre will be addressed in the science
assessment of the protocol.

	In response to a question from Dr. Fisher about the 5x5 design, Mr.
Carley explained that the AaiH stratum would differ for each MU within
the cluster.  Each worker in the cluster will handle a different amount
of pesticide.  The strata for AaiH are defined the same way for
all crops.  When data are collected for the other three scenarios, five
MUs per scenario in five different states will be collected.  This will
result in five data points for each AaiH strata, but the actual values
of AaiH for the five MUs in a stratum will vary.  For example, there
will be 25 data points across five strata and five data points for
application to pecans.

	Dr. Fisher asked if, after completion of the five field studies in this
scenario, the data within an AaiH stratum could be analyzed by combining
data from the different crop types.  Mr. Jordan explained that for each
of the 5 MUs, unit exposure will be calculated for each MU adjusted for
AaiH.  This will result in 25 normalized unit exposure values.  In the
past, EPA determined the 50th percentile of exposure and the geometric
mean for these data points.  The unit exposure will vary from person to
person based on work habits, so there will be a distribution of 25 data
points.  Based on the type of risk assessment activity EPA is
performing, the Agency will use the most appropriate unit value of
exposure based on duration of exposure.  EPA probably will analyze the
differences between crop types to determine if crop type has an effect
on exposure, given the different types of equipment and sprayer behavior
associated with each crop type.  In general, EPA will use the central
tendency value for multi-day exposure and the high end of exposure for a
single day for risk assessment activities.  Data points that generate
central tendency values are normalized exposure values from all 25
subjects.  Dr. Fisher noted that because there are no replicates, it
will be impossible to determine if exposure differences are related to
individual differences.  She recalled previous Board objections to
reliance on a single subject to represent all data points.

	Dr. Carriquiry added that EPA’s plans for using these data are
correct.  EPA is analyzing exposure of workers who apply pesticides
using a certain piece of equipment for different types of crops under a
variety of weather conditions.  There will be 25 data points that refer
to closed-cab airblast pesticide application for fruit crops.  This can
be thought of as drawn from a distribution of all closed-cab airblast
workers.  The data will give an idea of the distribution of exposures of
these workers.  Estimates of central tendency or means are appropriate,
but exposure will be overestimated for the 95 percentile of exposure;
this is acceptable because it is likely to result in more protective
conclusions about exposure.  The lower percentiles of exposures will be
underestimated, but this is not a significant concern.  EPA’s approach
for their intended use of this data is reasonable.

	Dr. Fisher asked how dangerous exposure levels would be determined from
this data.  Mr. Jordan answered that this would be discussed in the
science assessment of the protocol as well as how values derived from
this data set fit into calculations of overall exposure.  The unit
exposure a worker receives when handling a pesticide is one component of
exposure, and the Agency will explain an algorithm that it will use to
determine upper-band estimates of high exposures.  Toxicology data will
be used to determine how exposure compares to levels that are not
expected to have an adverse effect.  Dr. Linda Young noted that
proportionality of AaiH to exposure appears to be a key assumption for
these activities and asked if this relationship had been validated.  Mr.
Jordan answered that EPA has a number of studies that permit examination
of the relationship between AaiH and unit exposure.  In some studies,
there is strong support for a proportional relationship, but other
studies are inconclusive and some show no proportional relationship. 
EPA is asking the AHETF to examine as a secondary objective the
proportionality assumption for each scenario.  Thus, they designed the
study to incorporate stratification of AaiH.  This will generate a data
set with a useful spread of data points from which to examine
proportionality.  If the relationship is not proportional, EPA will
attempt to determine factors that drive variability in unit exposure. 
Dr. Carriquiry stated that because EPA will have data on exposure and
amount applied, they could determine distribution normalized in a
different way if the proportionality assumption does not hold.

EPA Science Assessment of Field Study Protocols AHE55 and AHE56

	Mr. Jeffrey Evans (OPP, EPA) presented EPA’s science review of AHETF
protocols AHE55 and AHE56.  These protocols are designed to develop
dermal and inhalation contact factors to assess potential exposure of
individuals making airblast applications to orchards and trellis crops
using enclosed cab vehicles.  The results will be used to populate the
closed-cab airblast scenario (a total of 25 monitoring units) in the
PHED database.  Protocol AHE55 will assess exposure occurring during
application of pesticides to citrus crops (five MUs) and protocol AHE56
will assess exposure during application to pecan crops (five MUs). 
Citrus and pecans represent taller orchard canopy crops.  

	The sites selected for AHE55 are located in Polk and Hillsborough
counties in central Florida.  There are approximately 7,000 citrus farms
in Florida and about 2,500 in Polk and Hillsborough counties.  The local
experts consulted suggested that enclosed cabs are widely used for
pesticide application in these counties.  The surrogate pesticides from
which participating growers can choose are Carbaryl and Malathion;
between 50,000 and 90,000 acres of Florida citrus are treated with one
of these chemicals each year.

	The sites selected for AHE56 are located in Tift County and surrounding
counties in southern Georgia.  Pecans are grown throughout Georgia on
approximately 3,700 farms covering roughly 128,000 acres.  Approximately
26,000 acres of pecans are grown in the study area.  Experts consulted
suggested that enclosed cabs are widely used in this area.  As with
AHE55, the surrogate pesticide options are Carbaryl and Malathion and
10,000 to 15,000 acres of Georgia pecans are treated with one of these
chemicals each year.

	Although there are only two surrogate AIs, they may occur in any of the
following formulations:  

Carbaryl-containing products:  Sevin® Brand XLR Plus Carbaryl
Insecticide (EPA Reg No 264-333), Sevin® Brand 80WSP Carbaryl
Insecticide (EPA Reg No 264-526), and Sevin® Brand 4F Carbaryl
Insecticide (EPA Reg No 264-349).  

Malathion-containing products:  Gowan Malathion 8 (EPA Reg No 10163-21),
Malathion 8-E Insecticide (EPA Reg No 34704-452), Fyfanon® (EPA Reg No
5905-196), and Fyfanon® 8 lb. Emulsion (EPA Reg No 5905-250-ZA).

	Each of the five MUs will apply a different amount of AI per monitoring
event based on the following strata:  5 to 9 pounds, 10 to 17 pounds, 18
to 30 pounds, 31 to 55 pounds, and 56 to 100 pounds.  EPA considers
these strata to be appropriate for airblast application and are
consistent with amounts used in studies to generate data for the PHED
and with amounts used in EPA handler exposure assessments.  The lower
limit and minimum of three tank-loads are likely to result in measurable
residues on all sample matrices, minimizing potential non-detects. 
Normalized data (mg exposure per pound AI handled) can be used to model
longer work days or higher application rates than the proposed study
durations (approximately 4 to 8 hours).  EPA wished to obtain
measurements on individuals performing these application activities; the
data will then be used to assess exposure of people handling more or
less pesticide.  This relies on the assumption of proportionality
between mg exposure per pound AI handled.  The protocols also will note
how much product workers apply and their behaviors.  Ultimately, there
will be 25 unit exposure values for this scenario.

	In estimating handler exposure, EPA assumes the maximum permissible
application rate, the maximum practical acres treated per day, and a
conservative estimate of dermal absorption.  The dermal exposure
algorithm will determine dose as a function of unit exposure times
application rate times acres per day times percent dermal adsorption all
divided by body weight.  The margin of exposure (MOE) will be calculated
as the no observable adverse effect level (NOAEL) (mg per kilograms per
day (kg/day)) divided by the dose calculated as described above. 
Historically, a standardized body weight of 70 kg has been used in these
calculations, but this value may be reconsidered.  The value for percent
dermal adsorption will be derived from animal studies and assumed to be
the same for all workers.  The goal is to achieve a MOE of 100.  

	Dr. Carriquiry raised questions about the MOE calculation.  She noted
that the assumptions about dermal absorption and body weight, and that
the acres treated per day are based on the number of acres a person and
machine could cover in 8 hours of spraying and conversion of this amount
to determine the amount of pesticide handled.  She requested
clarification of the MOE given that there will be 25 different values
for exposure, based on MU use of different strata of AaiH.  Mr. Evans
explained that to estimate MOE, 25 unit exposures could be used, or EPA
could perform a point estimate or select an arithmetic mean.  If a
probabilistic approach is used, each value could be a distribution from
which EPA could select.  Mr. Jordan explained that the application rate
is a product-specific value that varies based on pesticide type.  EPA
likely will choose the highest allowed application rate for its risk
assessment.  Application rate and number of acres treated are the
upper-bound values.  Percent absorption is a conservative estimate. 
This will result in a high-end estimate of a worker’s potential dose. 
Dr. Prentice inquired if EPA would measure variability in application
rates across workers, or just assume a maximum rate.  Mr. Jordan replied
that EPA would assume a maximum application rate, although the Agency
recognizes that less likely will be applied.  EPA makes these
assumptions to assure safe exposure at the high ends of application. 
Dr. Young asked if the difference between the high end and low end of
unit exposure was based on having a higher concentration of pesticide in
the sprayer tank or by spraying for longer time periods.  Mr. Jordan
responded that worker behavior, such as exiting the cab to unclog a
spray nozzle, was more likely to affect exposure.  Mr. Carley clarified
that the differences in product handled with respect to the AaiH strata
is primarily based on the size of the sprayer tank and the size of the
orchard; concentrations of the pesticides are roughly consistent.  The
variables include the size of the tank, speed of the truck, and acreage;
these variables are considered during the design phase when the growers
and the strata for which they are candidates are selected.

	Dr. Johnson asked about the relationship of the MOE to product
registration, given the assumption that EPA will use the data for risk
assessment activities to inform registration of products.  Mr. Carley
replied that the target MOE is at least 100.  The predicted high end of
exposure is based on the conservative assumption that it should be no
more than 1/100th of the NOAEL.  EPA also might use the data to inform
calculations for open cab application.  EPA would assume a MOE of only
25 for this type of application; if the exposure data supports this MOE,
EPA can require airblast use of these pesticides only when enclosed cabs
are used.  These data also only apply to spraying; data from other
behaviors such as mixing and loading also will have to be considered in
the risk assessment calculations.  EPA also could require a specific,
lower acceptable application rate.  This research is based on the idea
that exposure potential can be modeled using surrogate data, i.e.,
individual values for each of the 25 MUs.  EPA hopes to be able to
measure exposure for any pesticide and for any crop using this method. 
EPA would not usually use data from each participant separately, but is
attempting to improve its deterministic assessment.  Dr. Chambers
inquired if the data would be used to predict individual exposure.  Mr.
Carley responded that the data would not be used for this purpose.  The
research models high ends of exposure to protect users while applying
these pesticides using closed-cab airblast in orchards.  EPA is
attempting to determine the worst-case exposure possibilities and cap
potential exposure using appropriate regulatory controls.

	Mr. Evans continued his science review of AHE55 and AHE56.  The AHETF
will collect dermal and inhalation exposure measurements to develop unit
exposure or other exposure metrics for both protocols.  The Task Force
recognizes that normalizing factors other than AaiH may be more
appropriate for applicators using closed-cab vehicles.  The resulting
data can be used to estimate dermal and inhalation exposure to the
monitored surrogate pesticides and to other pesticides registered for
closed-cab airblast application to orchard or trellis crops.  To
determine inhalation exposure, airborne concentrations of the surrogate
will be monitored in the subject’s breathing zone using an
Occupational Safety and Health Administration Versatile Sampler tube
sample collector connected to a personal sampling pump.  Particle size
will not be measured; all residues will be assumed to be respirable. 
Dermal exposure will be measured by analyzing a 100-percent cotton whole
body dosimeter (long underwear) worn beneath the subject’s clothing. 
This will be performed in accordance with the draft EPA Series 875 Group
A–Applicator Monitoring Test Guidelines.  Measurements on the hands,
face, and neck will be collected by rinse and wipe methods using 0.01
percent Aerosol OT solution according to OECD Applicator Guidelines. 
Foot exposure is expected to be insignificant, as determined by open cab
data on foot exposure in PHED and thus socks will not be collected
as dosimeters.

	For hand wash and face/neck wipe samples, EPA proposes the following
conditions:  

if measured exposures from hands, face, and neck contribute less than 20
percent of total exposure, no action is required

if measured exposure contribution represents between 20 and 60 percent
of total exposure, an automatic 50 percent adjustment will be made or
the AHETF can submit a validation study

if measured exposure contribution is greater than 60 percent, a
validation study is required.  

Existing hand rinse data for Malathion exposure in field workers
indicates 59 percent removal in the first rinse; combined with a second
rinse, a mean of 85 percent removal would occur.  These data are derived
from validation studies involving intentional dosing of human subjects,
and therefore require prior review by EPA and the HSRB.

	These protocols address the technical aspects of applicable exposure
monitoring guidelines.  EPA Series 875 Group A–Applicator Monitoring
Test Guidelines, OECD Applicator Guidelines, and Good Laboratory
Practices [GLP] (40 CFR part 160) have been met.  In its science review,
EPA has concluded that while generally acceptable, the protocols could
be improved slightly.  As a contractor to the AHETF, the LSC is a member
of the research team and therefore should not also serve as part of the
study population, either as an expert or as a cooperating grower or
commercial applicator.  The processes of diversity selection of growers
and construction of an efficient configuration of MUs should be fully
documented in an appropriate SOP.  Finally, the numbers of growers and
workers in the working pool should be increased so that individual
workers filling all MU slots can be selected randomly from among
qualified, interested volunteers.  In its communication to EPA dated
June 13, 2008, the AHETF agreed to make these changes.

	Clarifying Questions

Dr. Lebowitz requested clarification concerning how samples would be
collected and stored, and asked about time to analysis and the methods
of analysis.  Mr. Evans answered that the governing documents discuss
use of controls such as travel spikes and field fortifications, as well
as quality assurance (QA)/quality control (QC) protocols and limits of
detection.  Field fortifications collected at the time of the studies
and travel spikes will be used to control for pesticide degradation. 
The Task Force recognizes that potential degradation of pesticide
residues is an important matter.  The field fortifications will contain
a known amount of pesticide; if decay is observed, the dosimeters can be
adjusted to account for loss.  In general, recovery from the dosimeters
is efficient, the samples are stored on ice, and residues are
successfully recovered.

Dr. Fenske noted that one point of the study was described as generating
data that will allow MOE and exposure analysis of pesticides other than
those used in this research; however, compounds of higher toxicity will
have different NOAELs but the same dose and application rate, which
would result in different MOEs.  Protective clothing requirements vary
according to the toxicity of the compound and this research uses lower
toxicity pesticides.  He asked how use of different types of PPE would
be factored into the equation when the data are used to assess exposure
to pesticides of higher toxicity.  Mr. Evans explained that the cab is
an engineering control; applicators wear what is required based for use
of the pesticide outside of the cab.  One issue that has arisen is
whether respirators are required, because few manufacturers will certify
that a cab is completely closed.  Thus, regulations often indicate that
respirators must be worn in the cab.  An estimate of exposure in the cab
when not wearing a respirator will be obtained from this research, and
then a respirator protective factor can be applied.  Dr. Fenske
countered that a person performing airblast application from an
air-conditioned cab likely will not exit the cab often, and thus it
would be surprising if exposure is detected on the inner dosimeter
garment.  Outside of the cab, workers are supposed to wear protective
clothing according to label requirements.  If a worker is spraying a
more toxic compound, he may be required to wear a coverall, which means
that the NOAEL and PPE requirement are different.  Mr. Evans explained
that PPE requirements for pesticides generally are based on acute
toxicity information.  For risk assessment activities, EPA will examine
data from a database of higher-toxicity compounds and evaluate if the
PPE is sufficient.  A protection factor for when a worker is wearing
coveralls can be assumed.  Dermal exposure measurements likely will be
small; however, it is unclear if the risk management decision to require
coveralls is useful.  Dr. Fenske stated that the dose value in the
denominator of the equation calculating MOE (NOAEL/dose) will be the
same, because it is based on workers wearing long-sleeved shirts and
pants.  If the same dose is used for higher toxicity compounds, the MOE
will be lower but will have no value for chemical protection required by
the label.  Mr. Evans agreed, and noted that the data could be adjusted
for wearing coveralls.  

Dr. Fenske inquired if dermal exposure as determined by hand and
face/neck washes would be collected for each worker.  Mr. Evans
responded that these values would be collected for each MU.  Dr.
Carriquiry questioned whether dermal exposure measured for working in
closed cabs was expected to be low, and if so, how this could be
proportional to AaiH.  Mr. Evans replied that one goal of the research
was to assess this relationship.  The researchers are aware that other
behaviors may affect exposure.  By analyzing measurements taken during
the study along with existing data, it should be possible to
characterize this relationship.  Dr. Fenske noted that the probability
that exposure will be based on AaiH and be consistent is fairly low. 
This will be a particular issue for this closed cab scenario.  Mr. Evans
agreed and noted that EPA has recognized this potential problem.  Dr.
Carriquiry inquired how the design would be affected if exposure is
found to be independent of AaiH.  Mr. Evans answered that EPA will need
to determine whether it is still a reasonable assumption to assume a
proportional relationship from a regulatory perspective.  Other
variables could be used to normalize exposure, but this will be
difficult.

In response to Dr. Fisher’s question about exposure occurring when the
worker exits the cab, Mr. Evans explained that behaviors exhibited and
time spent outside of the cab will be reported; however, there will be
no controls for this aspect of the study.  Dr. Fisher noted that because
any differences in exposure will occur when the worker is outside of the
cab, and because the design has no control for this, each data point
will be “independent” and the usefulness of the data is
questionable.  Dr. Carriquiry commented that the policy regarding
spraying three tanks of pesticide to determine the amount applied, and
how this relates to the design and AaiH strata is unclear.  It seems
likely that a person applying 100 pounds of pesticide would exit the cab
more frequently, which could cause confounding between the AaiH and the
time to apply the AI.  The time spent applying pesticide is more likely
to be proportionally related to exposure than AaiH.  

Dr. Susan Fish requested clarification of the utility of the data.  It
appears that this 5x5 configuration, with 25 different situations has a
number of potential confounders, which may result in the data resembling
a “case series” of 25 individuals, i.e., a qualitative study of each
particular MU.  This data would be qualitatively useful, but not
quantitatively useful, and impossible to statistically manipulate.  Dr.
Fenske explained that, in EPA’s view, the research will generate 25
high-quality measurements of a time period of exposure under the general
condition of closed-cab airblast application.  There are thousands of
worker-days for application of pesticides using this method per season;
if 1,000 workers were sampled, the same level of diversity would be
observed as with the 25 workers.  The hope is that these 25 workers will
be a sample of total worker-days that will allow EPA to calculate the
upper bound, 95th percentile of the distribution of exposure.  The data
likely will be better than existing data used for risk assessment
activities.

Dr. Fisher noted that although the MOE calculation is understandable, it
is difficult to understand how data generated for each of the MUs can be
analyzed statistically in any way, given the high degree of variability.
 Unless there is an assumption that different AaiH will be applied
across the five scenarios, there will be 25 different data points. 
Given the costs of this research, it is difficult to understand the
advantage of maximizing variability across five scenarios versus
replicability in fewer scenarios.  She also asked if there is a need for
five participants per AaiH strata, because the relevance of
measurements taken at lower dosage is questionable.  Another alternative
would be to include replicability to control for individual error, which
may be more important than variability across different crop types. 
This may also permit a stratum to be removed and thus costs would be the
same, but a better control for error would be possible.  Mr. Evans
explained that driving tractors through orchards is a basic activity. 
Obtaining measurements from many different people participating in many
different activities related to pesticide handling is important for
assessing exposure.  The available data show that measurements can be
made and the issue of non-detects can be resolved.  EPA needs to address
the issue of magnitude of exposure for workers using closed cab
tractors.  Mr. Crowley agreed that replicability would be desirable, but
it would lead to an exponential increase in the number of samples needed
to capture variability.  It is unclear, from a regulatory perspective,
if a limited data set with smaller error would be better than the
current approach, or if data from a limited set (i.e., single crop type)
could be extrapolated to other scenarios.

Dr. Fish noted that there are many unknowns for this research.  She
asked if it would be possible for the HSRB to review data from a
completed portion of the research and, based on the results, require the
AHETF to perform the research using additional AaiHs, scenarios, or
replicates.  Dr. Fisher noted that the AEATF used three replicates and
the HSRB had concerns about this, although the Board understood the need
for this number of replicates.  The Board stated that its approval of
the protocol at the time it was presented did not guarantee that the
data will be acceptable once the protocol has been executed.  The
Board’s interest is in ensuring that the research does not generate
un-interpretable data.

Dr. Young inquired if time spent outside the cab will be tracked.  Mr.
Carley replied that complete observations of activities, such as the
time it takes to spray the orchard, distance the worker is from mixing
and loading activities, and other parameters, would be made; however,
the activity of the workers in this research is not tightly scripted
because the goal is to incorporate “natural” variability that occurs
in the world of applicator days.  The variability observed among the 25
MUs is likely to be representative of the variability in the
agricultural world.  Applying pesticides is a highly variable activity,
and each application can be viewed as its own “story.”  The idea of
developing a schematic in which points of replicability can be chosen is
not possible, because some behaviors cannot be replicated.  The research
will generate data for a number of scenarios; EPA will analyze the data
and present it to the HSRB.  After this, EPA will determine if a
scenario has been sufficiently explicated to allow the Agency to assess
future closed-cab airblast applications.  Until the data are generated,
it is not known what the data will show.

Dr. Carriquiry agreed with Mr. Carley, but asked why, given that the
proportionality assumption had not been proven, a pilot study with one
grower and five MUs had not been proposed.  Mr. Carley explained that
EPA has data from approximately 15 studies, several of which were
performed using open cab airblast application and modern dosimetry. 
Based on this existing research, the AHETF is reasonably sure that
measurable residues will be detected.  If no relationship between AaiH
and exposure is found for the closed cab application, but is found for
the open cab application, these results are nonetheless informative,
from a regulatory perspective, about the ability of a closed cab to
mitigate risk.  If the research shows that exposure occurs when the
worker exits the cab, EPA will use that information to develop
strategies to reduce risk.  A strong correlation between AaiH and
exposure in the open cab scenario is expected, which would speak to the
ability of a closed cab to effectively protect against exposure.  The
amount of pesticide that deposits on the outside of the cab also might
impact exposure.

Dr. Fisher asked when data from this protocol would be available.  Mr.
Carley responded that the Board will see the data when the entire
scenario is populated with all five studies, the timing of which depends
on whether the other three scenarios are ready in time for the
October 2008 HSRB meeting.  If the Board finds the protocols presented
at this meeting to be acceptable, AHE56 will be executed in August 2008,
and AHE55 (and other protocols, if found acceptable) will be executed in
spring or summer of 2009.  Dr. Fisher asked that the Board be provided
with the prior open cab studies for informational purposes.  Mr. Carley
explained that EPA will meet with the AHETF shortly after this meeting
for further discussion.  The AHETF has sent preliminary documents for
the next round of studies that will complete both the open and closed
cab airblast application scenarios.  EPA hopes to bring these protocols
to the Board’s October 2008 meeting.

EPA Ethics Assessment of Field Study Protocols AHE55 and AHE56

	Mr. Carley presented EPA’s ethics review of AHE55 and AHE56. 
Assessment of the value of this data to society is based on the need for
exposure data for applicators using closed-cab airblast sprayers to
support EPA exposure assessments.  Few data currently exist, and these
data do not meet contemporary standards.  The knowledge likely to be
gained from this research will be usable in exposure assessments for the
specific crop uses and pesticides monitored as well as for other crop
uses and pesticides employing closed-cab airblast equipment.

The AHETF proposes to recruit subjects for these protocols from among
the employees of eligible growers in the “working pool.”  Screening
of growers will continue until the working pool includes growers with a
total of at least 10 workers who may potentially volunteer for the
study, and at least two workers for each defined AaiH stratum.  The pool
will include more growers and more workers than are ultimately needed
for the study.  In its letter to EPA dated June 13, 2008, the AHETF
agreed to “over-recruit” growers to provide at least three or four
potential workers who meet the criteria for each needed MU.  Subjects
will be recruited from among workers who are employed by eligible
growers in the working pool (or by pesticide application service
companies used by eligible growers in the working pool), have recent
experience using the specific closed-cab airblast application equipment
to be used in the study, and meet the subject eligibility requirements
of the study.  If more workers are available and interested than are
needed, subjects will be selected randomly. 

EPA has determined that the inclusion factors are appropriate. 
Exclusion factors also are appropriate, except for the exemption of
employees of the LSC from the exclusion that applies to employees of all
other AHETF contractors.  Appropriate steps have been proposed to
protect candidates and subjects from coercion or undue influence to
participate, again with the potential exception of employees of the LSC.
 The AHETF agreed to eliminate the exemption of LSC employees from the
exclusion group in its June 13, 2008 response to EPA’s review of the
protocols.  As per protocols of this type, growers must agree not to
coerce employees to participate, or to influence their decisions in any
way.  Employers of potential subjects also are excluded from all
recruiting meetings of the investigators and subjects.  EPA’s
overarching concerns in recruiting, screening, and obtaining consent can
be divided into four categories:  equitable subject selection, fully
informed choice, fully voluntary choice, and respect for subjects.

Matters related to equitable subject selection include appropriate
inclusion/exclusion factors, appropriate recruiting strategy, and
representativeness of the sample.  The proposed inclusion criteria are
appropriate and acceptable, and include the following:

willingness to participate and sign the Informed Consent Form (ICF)

handles pesticides as part of their job

trained in safe pesticide handling per Worker Protection Standards (WPS)

experience within the past year applying pesticides to the target crop
using the airblast sprayer and tractor to be used in the study

at least 18 years old with government-issued identification

in good general health

willing to follow all label and WPS requirements.

Exclusion criteria include being a pregnant or nursing mother, normally
wearing more PPE than is required by the label of the surrogate
pesticide, lack of understanding of English or Spanish, and employment
by a pesticide manufacturer or a contractor to the AHETF.  Procedural
exclusions include workers employed by growers who were unlisted or who
declined to participate; these exclusions arise because of the sampling
strategy developed for these protocols.  Exclusion of workers who
normally wear more PPE is necessary to avoid confounding the data; the
surrogate pesticides were chosen in part because they do not require
more protective clothing than long-sleeved shirts and long pants. 
Literacy is not required and appropriate provisions have been made for
informing and obtaining consent from workers who do not know how to
read.

Representativeness of the sample is achieved by thoughtful selection and
justification of purposive elements in the sampling design, and the
elements of random selection that were incorporated appropriately into
the sampling design.  The resulting sample is likely to include diverse
characteristics expected to affect exposure, with minimal selection bias
and be useful for modeling future exposures.

To ensure fully informed choice, the consent forms contain all elements
required by 40 CFR §26.1116 in accessible form and language.  The
consenting process is adequately described in the protocol.  Equivalent
processes for Spanish and English speakers, relying on bilingual
investigators, have been described; however, EPA raised concerns
regarding the use of interpreters, which has not yet been addressed by
the AHETF.  The SOPs indicate that if best efforts to obtain a bilingual
investigator fail, the AHETF reserves the right to use translators, but
how these translators will be identified and qualified has not been
documented.

To adequately communicate risks and benefits, the organization and
content of risk discussion in consent forms have been found to be
acceptable; however, the label and Material Safety Data Sheet (MSDS)
summaries cited in the SOP have not been approved by IRB, and may not be
available in Spanish.  EPA recommended that this SOP be revised to
delete this reference and include all necessary information in the
consent forms.  Benefits include knowledge about individual exposure
results that are presented in context.  EPA has advised that information
about an individual’s exposure related to the exposure of others be
included for comparison.  The AHETF has agreed to provide this context
as well as individual results, if requested.  The study director
currently is responsible for confirming subjects’ understanding;
however, this will be difficult if the worker speaks Spanish.  Appendix
3 to SOP 11-B provides example questions for confirming understanding.

Study candidates must speak English or Spanish, but are not required to
read; “impartial witnesses” will ensure that non-readers are fully
informed.  The SOP calls for a bilingual impartial witness, but how to
ensure that the witness is fully bilingual and impartial is not
explained.  If English is the candidate’s preferred language,
recruitment will be conducted in English by the principal investigator. 
If Spanish is the candidate’s preferred language, recruitment will be
conducted in Spanish by a bilingual member of research team.  The AHETF
has decided to include bilingual investigators in all research
protocols, which would obviate the need for translators.  

To manage dependent relationships, cooperating growers must promise
neutrality; employers are excluded from recruiting meetings, and
employees of interested entities (except the LSC) are excluded as
subjects.  To minimize peer pressure, the recruitment interview is
private.  Pregnancy testing is managed to maintain privacy.  Because of
the need to provide real alternatives to participation since the
research takes place at potential subjects’ work sites, if a worker
declines to participate, the research will be conducted elsewhere, with
another worker.  Candidates who attend an individual interview will be
paid $20.  Enrolled subjects who wear the dosimeter will be paid $80,
regardless of whether spraying occurs and the research can be conducted.
 Subjects are free to withdraw at any time, for any reason, individual
results will be made available by request, and medical care for
research-related injuries will be provided at no cost to the subjects. 
To ensure privacy and confidentiality, no records will be kept of
candidates who do not qualify or consent to participate; positive
results of pregnancy testing will discussed only with the individual
candidate, and records will be shredded; collected data will be
identified only by subject code; observations will not be shared with
subjects’ employers; linkage of subject names/addresses to subject
codes will be maintained securely; and subjects will not be identified
in reports or in databases.

EPA has concluded that the processes for recruiting, informing, and
obtaining consent from the worker-subjects of the proposed research,
with minor changes the AHETF has agreed to make, satisfactorily address
all of the Agency’s overarching concerns.

Risks and risk minimization activities fall into six categories: 1)
heat-related illness; 2) exposure to surrogate chemicals; 3) scripting
of field activities; 4) psychological risks; 5) exposure to detergents;
and 6) background risks of agricultural work.  To minimize risks of
heat-related illness, unhealthy subjects are excluded, tractor cabs are
air-conditioned, subjects are closely observed by research staff trained
to recognize signs and symptoms of heat-related illness, and medical
staff will be available onsite.  SOPs to address contingencies were
developed, and stopping rules based on subject activity and heat index
with thresholds lowered as recommended by the HSRB will be in place. 
Exposure to surrogate chemicals will be mitigated by using AIs that are
fully tested and well understood and by ensuring MOEs are substantially
greater than the target.  Because there will be little scripting of
field activities, scripting is unlikely to substantially affect worker
risk; however, subjects must be trained (or exempt) per WPS, and
experienced with the specific equipment to be used.  The degree of
scripting may vary with tank size, etc., to achieve the target AaiH. 
Psychological risks attributable to pregnancy testing and assisted
dressing or changing into the whole body dosimeters are minimized by
ensuring privacy of testing and results and by providing same-sex
assistance for dressing.  Risk from exposure to detergents present in
hand wipes or face/neck wipes is minimized by using highly diluted,
freshly mixed detergents.  The detergent concentrate is an eye irritant
and eye wash stations will be available in case of need.  Background
risks of agricultural work include those associated with operating the
tractor and airblast sprayer, and those associated with potential
tank-mix products which may be combined in the tank with the surrogate
pesticide.  Investigators will ensure that workers are familiar with the
equipment and that all label requirements for tank-mix products are
observed.  Additional products in the tank will not confound analysis of
the surrogate residues.

This research has no direct benefits to subjects.  Indirect benefits to
subjects include knowing their own exposure and how it compares to that
of others.  The societal benefit from improved data for risk assessment
is likely to be realized.  Growers benefit from free product but also
absorb the costs of inconvenience and lost time.  Sponsors benefit from
the lower cost of shared surrogate data development.  EPA considers the
risks associated with this research to have been fully identified and
effectively minimized.  Any residual risks to subjects will be low and
are reasonable in light of the potential societal benefits of obtaining
reliable data on dermal and inhalation exposure while applying liquid
pesticides to crops with airblast sprayers drawn by vehicles with
enclosed cabs.

Independent Investigational Review Board, Inc., of Plantation, Florida
(IIRB) reviewed and unanimously approved the protocol and English and
Spanish consent forms and product-specific supplements.  IIRB is
independent of the sponsors and investigators, is registered with the
Office for Human Research Protections (OHRP), but is not accredited. 
IIRB procedures have been submitted directly to EPA under a claim of
confidentiality; EPA has determined they meet regulatory standards. 
IIRB provided the Spanish translations and reviewed the product-specific
supplements to the consent forms.  This is a proposal for third-party
research involving intentional exposure of human subjects to a
pesticide, with the intention of submitting the resulting data to EPA
under the pesticide laws.  The primary ethical standards applicable to
this research are 40 CFR part 26, subparts K and L.

Deficiencies noted by EPA that require adjustment include identification
of the LSC in the protocol; justification or elimination of the
exemption of LSC employees from the general exclusion of employees of
AHETF contractors; ensuring that all materials provided to candidates in
the recruiting and interviewing processes, including label and MSDS
summaries and other materials described in SOPs, are IRB-approved before
use; and clarification of the use of the questions in Attachment 11-B-3
to SOP 11.B.1 to confirm candidate understanding.  The AHETF has agreed
to eliminate the exemption of LSC employees from the general exclusion
of employees of AHETF contractors.

EPA also requests that the AHETF define the standard of documentation
required to support a decision to rely on an interpreter rather than a
bilingual investigator, and procedures for how interpreters and
“impartial bilingual witnesses,” if required, will be recruited and
compensated, and how their competency and impartiality will be assessed.
 The AHETF claims they will rely on bilingual investigators, but the
SOPs and protocols provide for use of interpreters, and these issues are
of concern to EPA.  EPA also asks that the AHETF count and report the
number of potentially eligible workers linked to each grower, and the
numbers attending initial group meetings, attending individual consent
interviews, signing consent forms, subsequently withdrawing or being
withdrawn, and completing participation; the AHETF has agreed to provide
these numbers.

EPA has concluded that all requirements of 40 CFR §26.1111, §26.1116,
§26.1117, §26.1125, and §26.1203 are met.  If all noted deficiencies
are addressed, the AHETF Closed-Cab Airblast Scenario and Field Studies
AHE55 and AHE56 will likely meet the applicable requirements of 40 CFR
part 26, subparts K and L.

Charge Questions to the Board:

If AHETF’s proposed closed-cab airblast application scenario design,
field study protocols AHE55 and AHE56, and supporting documentation are
revised as suggested in EPA’s reviews: 

does the research appear likely to generate scientifically reliable
data, useful for assessing the exposure of handlers who apply liquid
pesticides using airblast equipment drawn by vehicles with closed cabs? 

does the research appear to meet the applicable requirements of 40 CFR
part 26, subparts K and L?

	Clarifying Questions

Dr. Fish noted that the supporting documents did not indicate how
literacy would be assessed.  Mr. Carley agreed that this information may
not have been included in the documents.  Dr. Jerry Menikoff questioned
whether the reimbursement for purchasing pesticides to be used in the
research could be considered an incentive for the growers.  He stated
that previous studies have shown that growers usually are not interested
in participating; however, if a grower with only one qualified employee
is interested in participating, that employee could be vulnerable to
coercion.  He asked how the AHETF would ensure that growers will not
learn that their employees did not consent to participate and if it was
possible, given the sampling strategy that growers could somehow
determine which employees declined to participate.  Mr. Carley answered
that such a situation would be more likely to occur if the AHETF
under-recruits growers.  If the working pool contains more than the
minimum required number of growers and at least three qualified workers,
the AHETF can choose workers randomly and inform growers that their
employee(s) was not selected.  EPA is satisfied that growers will not
learn the reason that they were not chosen for participation.

Dr. Rebecca Parkin inquired if workers must agree to be filmed while
applying the pesticide as a condition of participation.  Mr. Carley
responded that the AHETF would need to answer this question.  EPA
acknowledges that observations are needed, but has advised the AHETF to
minimize complications associated with observation.  EPA has recommended
that many observations could be done from a distance, which would help
mask the identity of the subject.  He agreed that the Task Force should
clarify the number of people observing the workers and also which
activities would be observed.  Dr. Parkin remarked that the ICF should
more adequately address this issue.  She also noted that no option was
given for having one’s face obscured if films or photographs will be
used for records or reports.  She expressed concern that this would
result in exclusion of subjects who wished to participate, but did not
wish to be recognized.  The AHETF also needs to consider storage of
these materials with respect to protection of identify.  Mr. Carley
replied that EPA intends to encourage the AHETF to include information
on this matter only in the SOP and not in the protocol, because
including the information in both places results in confusion.

Dr. Prentice asked if the participating workers would be seasonal
employees.  Mr. Carley explained that this would probably not be the
case.  Typically, very few workers apply pesticides, and these workers
are almost always permanent workers, or, in some cases, the growers
themselves.  This is a specialized task and only a few workers at each
site ever perform it.  Dr. Prentice asked if, given average salaries,
the $20 payment for the initial interview was high enough to be
considered coercion.  He also asked if, after completing the first
interview, there was any likelihood of coercion to continue further in
the process.  Mr. Carley responded that the consent form clearly states
that subjects are free to withdraw at any point.  Dr. Fisher commented
that the protocol should clarify that workers who attend the initial
interview receive $20 whether they choose to participate further or not.
 Mr. Carley added that there are no scripts for the initial meeting or
individual interviews, which is not unusual.  Dr. Menikoff noted that
the ICF does explain that workers attending the initial interview
receive $20 regardless of further participation.

Dr. Prentice inquired about the educational level of the subjects.  Mr.
Carley stated that members of the Task Force would need to answer that
question.  Dr. Carriquiry asked whether workers who apply pesticides are
required to be trained or certified.  Mr. Carley explained that
certification is required for those who supervise application of toxic
pesticides, which is regulated though a certification program.  The two
surrogates used in this research do not require certification, but the
eligibility criteria require that workers have received minimal training
as defined by the WPS.  All involved workers will be experienced
applicators, which implies that most participants will be able to read,
probably in English.  The full labels for the products could cover over
100 pages, and it is unlikely that all of this will be available in
Spanish.  Dr. Fenske countered that in Washington State approximately 80
percent of applicators cannot read the English labels.  Dr. Suzanne
Fitzpatrick asked if a grower could serve as a MU.  Mr. Carley replied
that this is possible and is provided for in the Task Force
documentation.  The exclusion criteria refer to those working in the
pesticide business or those who are employed by Task Force contractors.

Public Comments

	Dr. Victor M. Canez of BASF Corporation, Mr. Curt Lunchick of Bayer
Crop Science, Dr. Richard H. Collier of Landis International, and Dr.
Larry Holden of Sielken and Associates Consulting, Inc., all on behalf
of the AHETF

Dr. Richard Collier (Landis International), on behalf of the AHETF,
thanked EPA and the Board for the opportunity to provide comments.  He
proceeded to answer questions that arose during the review of the AHETF
protocols.

Dr. Collier explained that AHETF’s position is that the LSC is best
viewed as a member of the research team and the SOPs and documentation
will be changed to reflect this.  Thus, the LSC and his or her employees
will not be eligible to serve as growers or MUs.  This change in
designation and accompanying clarification will address a number of
scientific and ethical issues raised by EPA.

Regarding the numbers of workers per grower, two workers per grower or
commercial applicator will be placed in the working pool but only one
worker per grower will be selected for participation.  With this
limitation, most application equipment would be used only once.  More
than one worker per grower will be used only if the grower has more than
one applicator and sufficient acreage to avoid application of the
pesticide twice to the same part of the orchard.

Based on studies of the surrogates to be used in this research and the
sensitivity of the methods used to analyze their residues, the AHETF is
convinced that measurable residues will be detected in most or all of
the samples even for the lowest stratum (5 to 9 pounds) of AaiH;
however, if non-detects do occur, the sensitivity of the analysis means
that the data will still be useful for risk assessment.  The non-detect
level for previous research was significantly higher than for this
research; the analytical method used in this research has a stable
matrix and high sensitivity.

The AHETF agrees that county extension agents will be useful sources of
information about growers.  The list of resources provided by the AHETF
was not a priority ranking.  Extension agents will be used as a source
of information.

Workers who volunteer to apply the pesticide will not be involved in
mixing and loading duties, but the AHETF will not control the degree to
which they are in contact with these activities.  The Task Force wants a
minimal amount of scripting; the goal is to observe normal applicator
behavior and document it in detail.

The AHETF will reimburse growers for their purchase of the pesticide,
but the grower will obtain the product from a local distributor.  This
decision was made based on logistical matters; it is more
straightforward for each grower to purchase the products than for the
Task Force to distribute them.

Concerning historically low response rates for participation in
agricultural research, this is based on the experiences of Task Force
members who have performed such research for a number of years.  This
matter is not documented now, but will be in the future.  The degree to
which a responder indicates a reason for non-participation will be
tracked.

The AHETF plans to use one observer per applicator; members of the
research team present at the site will also include the study director,
a medical professional, and members of the team who will take the field
fortification samples at the site.  The Task Force expects that the
observer will be located at the edge of the field while observing the
application process and recording observations.  The AHETF recognizes
that the presence of an observer influences behavior.

SOPs are available that describe the storage, transport, and analysis of
the samples in great detail.  Samples from the dosimeters and the hand
and face/neck washes are frozen onsite with a corresponding field
fortification sample.  Analysis of the samples may take place some time
later.  The surrogates chosen for this research are known to be stable
under these collection and transport conditions.

The AHETF, as part of a regulated industry, recognizes the importance of
the issue concerning the proportional relationship between AaiH and
exposure.  Even in a low-exposure scenario, pesticides are regulated
based on this proportionality assumption; therefore, it was important to
include in this scenario design stratification by AaiH to generate data
to test this proportionality assumption.  Even if the degree of
proportionality is uncertain, measures of absolute exposure may be low
enough that the question of proportionality becomes irrelevant for this
particular scenario.

Regarding the use of bilingual investigators and interpreters, the AHETF
agrees with EPA’s concerns and has decided to always include bilingual
investigators when Spanish-speaking workers volunteer to participate. 
Interpreters will not be used; therefore, bilingual witnesses will not
be needed.  The QA officer will ensure that all aspects of the protocol
are followed and will document any deviations.

Dr. Victor Canez (BASF Corporation), on behalf of the AHETF, explained
that there is a high probability that more small growers will populate
the lower strata of AaiH, but over-recruiting should provide some
variation in grower size for these strata.  Regarding schedules for
application, the Task Force will ask growers when they intend to apply
pesticides.  Most growers have scheduled applications based on the
presence of certain pests.  The rate of application can be adjusted
slightly, but applications at non-efficacious rates, according to label
instructions, will not occur.

Dr. Young questioned whether small growers would be excluded if the
applicator also usually performs mixing and loading duties.  Dr. Canez
responded that in cases where one worker performs all these duties, a
member of the research team, such as the LSC, will perform the mixing
and loading.  Dr. Chambers inquired whether having the Task Force
provide the pesticides would result in better control of the quality
and age of the pesticides.  Mr. Curt Lunchick (Bayer Crop Science), on
behalf of the AHETF, replied that the Task Force will record lot numbers
and perform GLP characterization of the pesticides to determine the
amount of AI in the product.  Dr. Canez added that this approach also
provides growers with the opportunity to use their preferred pesticide. 
Dr. Lebowitz requested clarification concerning accidental or other
exposures of the MU during mixing or loading that may complicate
exposure measurements.  He also asked if the Task Force SOPs stated a
specific time period in which samples must be analyzed.  Dr. Canez
explained that no time period had been specified, but the Task Force
believes that the stability of the samples and the QC measures in place
are sufficient to ensure that deterioration will not be a significant
issue.  He added that the samples will probably be analyzed less than 90
days after they are collected, but this has not been specified.

Dr. Johnson requested clarification of the identities of the LSCs.  Dr.
Canez explained that having a LSC usually is a GLP or the LSC usually is
an agricultural researcher located in a particular area.  The LSC will
be familiar with the local farmers and their growing practices, crops,
and the chemicals they use.  LSCs may operate in several states; for
example, one LSC works in Florida, Georgia, and South Carolina.  The
Task Force intends to employ one LSC per area.  LSCs have been
identified for AHE56 and the Task Force is in contact with several LSCs
in California.  Once the LSCs are identified, they will be specified in
amendments to the protocol.

Dr. Krishnan inquired about the similarity of the application rates of
the surrogates.  Dr. Canez responded that the amounts of AI applied for
each surrogate are probably similar.  

Dr. Gary Chadwick asked about the number of different types of closed
cab tractors and airblasters among the 25 MUs.  Dr. Canez explained that
there are likely to be 25 different kinds of equipment.  The equipment
is similar in construction; most airblast sprayers spray the pesticide
upward or outward.  The closed cab tractors also will differ in type,
age, and upkeep, but the AHETF has specified that the cabs must have
functional air conditioning and closed windows.  In addition, growers
often modify nozzle positions, fan outputs, and other aspects of the
sprayers.  

Dr. Carriquiry stated that the 5 percent response rate expected by the
AHETF is quite low.  In her research, use of trained interviewers
usually results in a high response rate.  She encouraged the Task Force
to work with people trained to make initial contacts; such people can be
found at land grant colleges in any state and will probably have more
experience in these matters and thus be more successful than the LSCs in
recruiting growers.

Regarding sample size and the five MUs, Dr. Carriquiry explained that if
the need for five MUs was calculated under the assumption that all five
will work for different growers, five MUs will not be sufficient if two
workers can be selected from a single grower.  Dr. Larry Holden
(Sielken and Associates Consulting, Inc.), on behalf of the AHETF,
explained that the assumption was not that there would be five workers
and five growers.  The assumption was that a random sample would be
drawn from within a cluster; two MUs could be selected from the same
grower by chance.  He acknowledged that Dr. Carriquiry was correct that
if “sub-clustering” or grouping within the cluster occurred, there
is a potential for a variance component to exist.  Analysis at 25
different sites would be logistically difficult; grouping into five
“clusters” was done to contain costs and because the Task Force
believed it would be difficult to recruit 25 growers.  The Task Force is
aware that there would be one variance component and that there would be
structure beneath that and so tried to make each MU as diverse as
possible.  When the data are analyzed, any structuring will be
considered.  In addition, there may be two MUs from a single grower,
but there will probably be differences in the way each MU handles the
sprayer equipment.  Dr. Johnson inquired if it would increase costs to
have one MU per grower rather than include the possibility of having two
MUs for some growers.  Dr. Canez explained that this would increase
monitoring time and thus increase costs by approximately 50 percent;
this would include salaries for staff and other logistical issues.  The
optimal approach was to be at a site for one day to set up and take 3
days to perform the research.  Increasing the time needed could add
approximately $5,000 per day.

Dr. Young questioned if two MUs at the same grower would be spraying at
the same time.  Dr. Canez confirmed that this was possible if the grower
had sufficient equipment and acreage.  Dr. Young asked if this meant
that more observers would be needed at the site.  She also questioned
the increased costs for monitoring at additional sites, given that the
growers are not located far from one another in Florida.  Dr. Canez
explained that the research days are long because they include setting
up, dressing the participants, 8 hours of spraying, undressing the
participants, and processing the samples.  The researchers may not be
able to leave for the next site until 11:00 p.m.  Mr. Lunchick added
that two MUs at one farm could be monitored on two consecutive days,
but if the design is changed to one MU per farm, this might space out
data collection over a week.  In addition, if the next spraying is not
scheduled for some time, this also could increase costs.

Dr. Fenske inquired how the AHETF would address variations in worker
behavior.  For example, although the study materials state that the
windows of the cab must remain closed, a worker might open the window. 
The instructions to the observer state that the observer will report
unauthorized activities to the study director.  He asked if anyone would
intervene if unauthorized behavior is observed.  Dr. Canez explained
that this would be treated in the same ways as refusing to wear PPE,
i.e., data would not be collected from that MU.

Board Discussion

Scientific Considerations – Protocols AHE55 and AHE56

Dr. Fenske recommended that the reference to “liquid” pesticides in
the Charge Questions be changed to “pesticides applied as water
aerosols.”  He stated that the research is likely to generate
scientifically sound data that will be useful for EPA risk assessment
activities.  He commended the Agency’s evaluation and presentation of
the materials and the Task Force for addressing most of EPA’s concerns
in its June 13, 2008 letter.

Specific issues of concern have been largely allayed.  On the topic of
AaiH, Dr. Fenske expressed some concern that the 5 to 9 pound stratum
would be too low to generate detectable residues, but experience of the
investigators involved in this research seems to indicate that this
amount will be adequate.  Dr. Fenske also was concerned that the highest
stratum of AaiH would not be high enough, but a recent study of 35
applicators working over multiple days and performing 150 applications
found that the upper limit of AaiH was approximately 100 pounds.  He
agreed with AHETF’s decision not to examine higher levels of AaiH.

Dr. Fenske stated that he had been concerned about the vagueness of
contacting local resources, but the documents related to the research
have clarified this issue.  He also was concerned that small growers
would be excluded, given some of the language in the governing
documents, but the random listing of growers according to size might
alleviate this issue.  Applicators working on small farms might have
more opportunities for exposure and therefore should not be
under-sampled.

The largest area of concern is the behavior of the applicators.  Dr.
Fenske remarked that in his experience in such an application situation,
worker behavior will be a significant determining variable of exposure,
so he had concerns about how “naturally” workers would behave in the
research setting.  Moving from task to task provides opportunities for
exposure; therefore, a worker responsible for more activities will have
increased exposure, but such exposure cannot be studied in this scenario
because mixing, loading, and handling will be studied separately.  This
approach may underestimate the exposure of those who normally perform
all three tasks.  The Task Force notes that exiting and contacting the
surface of the cab may be the most significant route of exposure in this
scenario, but exiting the cab is usually a rare event.  A worker could
remain in the cab for the entire application duration and also would not
be exposed to the mixing and loading activities; Dr. Fenske recommended
that workers should make 0 or 1 exit from the cab or the data should be
discounted because other behavior would not be realistic.  He commented
that normal behaviors are constrained in the protocols because the
research will be stopped if a worker opens the cab window or takes off
his gloves, for example.  Workers often engage in activities that are
normally prohibited and thus are exposed in this manner; however, terms
such as “sloppy worker” should be avoided because workers often have
reasons for failing to comply with all protective regulations.  Because
of these constraints, the data will exclude exposures at the high end. 
This is of concern and EPA may have to consider how to account for
behavior in its use of the data.

Dr. Fenske described other minor concerns.  The documents cite
agricultural experts to support certain decisions, but actual citations
are not documented.  The Task Force should identify the experts with
whom it consults and provide references when possible.  Although the
documents indicate that only 15 percent of pesticide applications are
made to dormant crops, this represents one in six applications and it is
unclear why these were excluded.  The AHETF also might want to
reconsider the exclusion of other orchard types because of low acreage. 
The Task Force should also consider checking the nozzles on the
application equipment to ensure they are relatively consistent across
applications.  Regarding observation of workers, the instructions in the
materials state that there are necessary forms and notebooks for
observations, but these are not described in the SOP; the forms and
observation notebooks should be described in the SOP.  Dr. Fenske
commented that in his review of the materials he found only the English
version of the recruitment form, which had a statement asking volunteers
to contact Dr. Larry Smith in Ohio; it may be unreasonable to expect
workers in Florida to contact someone in Ohio if they wish to volunteer
for the protocol.

Dr. Fitzpatrick stated that the protocol was well thought out and
adequately answered the science charge question.  She asked if Mr.
Carley’s request for a SOP describing randomization of growers would
include a description of how MUs are chosen at random.  Mr. Carley
replied that he will request specification of the entire process,
including record keeping and reporting activity.

Dr. Parkin commented that the documentation for this research was
strong.  She requested clarification of how observations would be made
if spraying occurred at night or before sunrise.

Dr. Johnson agreed that the data would be scientifically reliable and
will be useful for assessing exposure.  He agreed with Dr. Edwards that
EPA needs this data for its risk assessment activities because existing
data are inadequate.  He commented that the influence of certain
parameters on activities needs to be understood with respect to the
variation in exposure, not only on the mean of exposure.  Conventional
sampling is inadequate because of the inability to obtain a random
sample of future conditions.  This research sought to generate
conditions to deliberately emulate realistic conditions, and the PDS
strategy seems to have achieved this goal.

Dr. Carriquiry stated that she was satisfied with the purposive
selection of states, crops, and counties.  She expressed some concern
about selection of growers; with some effort, sampling at this level can
be changed to result in a better, more valid design.  If the design is
valid, EPA is less likely to receive protests from the public or others
regarding regulations based on the data generated from this research.

The current plan for selecting growers incorporates random elements, but
has some major drawbacks.  If there is a high rate of refusal, the
current plan does not permit investigation or characterization of
non-responders.  This will impact the representativeness of the study,
especially because the sample is not nationally representative.  It will
be important to know if the 5 percent of growers who agree to
participate are representative of all growers; therefore, whether the
responders are similar to the non-responders with respect to size of
operation, types of equipment, number of employees, etc., needs to be
known.

The current plan also does not have a reasonable chance of including
small growers.  Small farmers tend to meet regulations less adequately
than do larger growers.  The Task Force thus should guarantee adequate
representation of growers of all sizes.

Dr. Carriquiry proposed an alternative design.  Given two Florida
counties with approximately 2,500 growers of known size, these growers
could be stratified by size (small, medium, large, or small and
medium/large).  From each size strata, a certain percentage of growers
(perhaps 2 percent) would be selected and an interviewer would contact
each grower to gather information about crop types, acreage, equipments
types, and number of employees.  This would create the sampling
subframe.  Of the growers in the subframe, five could be chosen randomly
within each size strata and then randomly allocated to AaiH strata.  The
advantage of this approach is that it excludes the possibility that
workers who refuse to participate are identified, and it permits
understanding of the characteristics of the non-responders.  Failure to
gather this information means the data will be useless if 95 percent of
growers do not respond.  This approach will guarantee the presence of
small, medium, and large growers in the dataset.

Mr. Carley inquired how MUs would be selected in this approach.  Dr.
Carriquiry responded that selection of growers by size will be
correlated with the number of workers.  Over-sampling will be needed
because the number of workers who will refuse to participate will be
unknown.  If five growers are selected from the different size strata,
three times as many workers could be chosen and contacted randomly.  The
main issue in the sampling design was selection of the growers; by using
this approach, the working pool will be larger and will help
characterize non-responders.  Dr. Chambers asked if workers could be
randomized from each size strata.  Dr. Carriquiry replied that this was
possible, but the main problem is that workers cannot participate unless
the grower does.  

Dr. Fisher noted that Dr. Carriquiry’s alternative model may raise
costs.  Dr. Young stated that the ability to obtain information on
non-responders is important.  She also expressed some disagreement with
Dr. Carriquiry’s model.  In Dr. Young’s experience, larger producers
are more likely to deviate from regulations than are small producers. 
Also, Dr. Carriquiry’s model could result in over-sampling of small
producers; however, limiting selection of MUs to one MU per grower would
be preferable.  Dr. Young added that being able to protect the identity
of growers and workers who participate is important.

Dr. Fisher noted that there appeared to be a strong preference for a
model that samples one MU per grower.  Other issues raised by Dr. Fenske
included confounding with respect to variability of the data related to
exiting the cab of the vehicle.  EPA should be given a recording sheet
that the observer will use that describes potentially affecting
variables, such as exiting the cab and proximity to the mixing and
loading activities.  Because exiting the cab is an important source of
exposure, the Task Force should consider including this variable as part
of the MOE equation.

Dr. Fenske stated that, except in some unusual situations, workers
generally exit the cab when the tank is being loaded.  This occurs at
fixed locations and should be easy to observe.  The observers will be
able to determine if the vehicle is moving, and workers will not exit
the cab while the vehicle is moving.  If the vehicle stops, the worker
may exit it.  It should be possible to gather useful data on worker
behaviors.

Dr. Fisher noted that the study was acceptable, as long as EPA is given
information on how behaviors will be observed.  Dr. Krishnan remarked
that as long as the data is not over-interpreted and the investigators
do not attempt to measure contributors to variability, the data can be
used to estimate dose for workers.  Dr. Johnson noted that although the
cabs are air-conditioned and the air intake is filtered, it is likely
that some exposure will nonetheless occur.  Dr. Lebowitz added that
pesticide is most likely to enter the cab when workers open the door or
track in residue on their shoes.  Given the amount of modeling occurring
after data are gathered, modeling exposure could be used to optimize the
design, particularly given that these data are intended for risk
assessment activities.  The AHETF’s review of the existing data in
PHED to inform their research plan should be commended.  Dr. Lebowitz
recommended that the Task Force continue to examine existing data on
pesticide exposure to further inform the design.  The EPA review of this
research and the Task Force documents contain information that justify
departure from the 5x5 design and raise questions about the suitability
of the PDS approach.

Dr. Fisher summarized the Board’s science review.  Overall, the Board
was impressed by the design of the research.  The Board questions the
usefulness of the data and cautions EPA against use of the data for
other than the intended purpose.  It is difficult to conclude that the
data will be “reliable” because of the lack of replicate measures;
however, the protocol will collect sound qualitative data that will be
useful to EPA.  EPA should consider doing the following:

ensure that workers from small and large farms are recruited

propose time limits for analysis of samples

obtain specific information about LSCs prior to implementation of the
research

ask the AHETF to employ experienced recruiters to increase the response
rate

clarify the wording of the charge question

obtain a form that describes how observational data, such as exiting the
vehicle, the quality of the machinery, and the MUs’ actions during
mixing and loading will be recorded

obtain information on how observations will be made if work occurs at
night

avoid terms like “sloppy” worker

provide citations in AHETF documents

suggest that the Task Force reconsider or explain exclusion of certain
dormant or other crop types.

The Board consensus is that selecting only one MU per grower would be
optimal.  It seems clear that attempting to allocate the 25 data points
in terms of farm size is possible.  If there are limitations to this
approach, the Task Force should articulate these, including increases in
costs, more precisely.  Continuing with the current plan, which limits
distribution, must be justified.  The AHETF appears to have a strong
analytic plan for the MOE, but exactly how this value is calculated is
not clear.  The Task Force should consider developing alternative
analytic plans in the event that the results of the research are
different than what is expected.  The Task Force also should consider
performing an initial survey to help predict the number of
non-responders.  If the Task Force surveys 50 growers, it can use this
information to determine the effects of non-response.  The AHETF should
also consider blinding the identity of growers and workers to EPA.

Mr. Carley clarified that the documents state that MUs will not be
identified.  The intent is to not identify growers, but regulatory
requirements for record retention may not permit complete
de-identification.  For example, if legal issues arose over matters
related to consent, EPA is allowed access to signatures and other
potential identifiers.

Ethical Considerations – Protocols AHE55 and AHE56

Dr. Fish opened the ethics review by stating that, with some changes,
the proposed research meets the applicable requirements of 40 CFR part
26, subparts K and L.  She then described several issues that should be
addressed.  The protocol does not include a description of how literacy
will be assessed; merely asking a participant if they are literate is
insufficient.  Spanish speakers should be available at all telephone
numbers that potential subjects are instructed to call for further
information, including those listed on the ICFs and for the IRB.  The
means by which subject understanding of the research will be assessed is
unclear; a quiz with suggested answers is provided in the appendix but
it is not clear if this will be used for all subjects.  The protocol
does not describe how persons conducting the consent interviews are
trained to assess subject understanding.  The background risk of
agricultural work should not be used when assessing the risk-benefit
relationship.  The Task Force should consider offering risk counseling
related to working with pesticides while pregnant to subjects who have a
positive pregnancy test; handouts or a referral would suffice.  The Task
Force also should consider obscuring participants’ identities in
photographs and video footage.  There is no discussion of the expected
gender distribution of applicators on the IIRB application.  The
predicted ethnicity distribution is 90 percent White and 10 percent
Hispanic in Florida and Georgia, which seems to be a low amount of
Hispanic participants.

The protocol states that the documents of those who withdraw from the
study because of a positive pregnancy test will be destroyed; EPA must
indicate whether this is acceptable.  References to interpreters and
witnesses should be removed from the documentation and SOPs for this
protocol because the Task Force has decided instead to use bilingual
researchers.  The ICF states that compensation for research-related
injuries will cover the cost of “reasonable and appropriate” medical
treatment; this phrase should be removed.  Dr. Fish agreed with
Mr. Carley’s suggestion to list certain procedures in the SOP rather
than the protocol because of inconsistencies between descriptions in the
SOP and the protocol.  SOP 11.F.0 for adverse event reporting states
that adverse events will be reported to Western IRB.  This languages
needs to be changed because adverse events should be reported to IIRB. 
The ICF refers only to the AHETF and not to the sponsor; at this point,
a re-write is not necessary but should be considered in the future.  The
Spanish translation of the recruitment poster states that subjects are
not qualified if they are “cognitively impaired;” this language
should be simplified.  If these changes are made, the protocol meets the
ethics charge question.

Dr. Chadwick noted that the section discussing use of photographs in the
ICF borders on exculpatory.  The document states that the AHETF owns the
rights to photographs and can use them for any purpose.  This asks
participants to waive rights that are unrelated to the research.  There
are ways to document the research using photographs that do not identify
the subjects.  The ICF should discuss the need for and use of the
photographs.  Dr. Menikoff agreed that the AHETF should clarify how the
photographs can and will be used.  Assuming appropriate limitations on
use, it is reasonable to not obscure identities at all times.  These
situations are not as embarrassing as typical medical photographs.  The
ICF states that subjects will not be photographed while dressing;
expanding the explanation of use of the photographs should be
sufficient.

Dr. Prentice inquired why the charge questions ask if the research
“appears” likely to generate reliable data and if EPA’s use of the
word “appear” in the charge questions was deliberate.  Mr. Jordan
explained that the use of “appear” was intended to convey that
despite the best plans, in practice the research may not comport with
standards.  EPA uses the charge question to determine that if a protocol
is performed as described, subjects will most likely be treated
ethically.  Dr. Fisher reminded Board members that the HSRB is not an
IRB.  The HSRB makes recommendations to EPA to ask for further
information or modifications to a protocol.  The protocol may return to
an IRB after Board review.

Dr. Prentice stated that he largely agreed with other Board reviewers
regarding the ethics charge questions.  He noted that there was some
confusion about when potential subjects receive a copy of the ICF.  It
appears that consenting may occur several days before or up to the day
of the study.  Volunteers should receive a copy of the ICF at least 1
day before the study.  Mr. Carley explained that subjects probably
receive the ICF at the initial group meeting and then can schedule
individual interviews; however, EPA will check that this is clarified. 
Dr. Prentice expressed concern about the payment of $20 to participants
who attend an interview, whether or not they participate in the
research.  Dr. Fish explained that this was usual practice at her
institution; the nominal payment is intended to cover the costs of
transportation and child or other dependent care.

Dr. Parkin noted that obscuring the identity of participants was
important not only for reasons of potential embarrassment but also to
protect undocumented participants.  She stated that she also was
uncomfortable with IIRB both translating and approving the translated
materials.  Dr. Menikoff asked if undocumented workers were a specific
concern, because the study design appears to make it unlikely that
undocumented workers will enroll in these studies.  Mr. Carley explained
that the anti-microbial exposure research taking place in California
required proof of age using a government-issued photograph
identification form that was not issued to illegal residents.  This
protocol also requires a government-issued form of identification and
thus illegal workers are unlikely to enroll.  Dr. Carriquiry noted that
the ICF does not specify that the identification must be issued by the
United States government.

Concerning translation by the IRB, Dr. Chadwick remarked that the IRB he
works with will obtain a translation of a document for an investigator,
but the process by which this is done was not known.  Dr. Fish explained
that commercial IRBs often have Spanish-speaking translators on staff. 
She added that using IRB staff members to translate documents is
preferable because they understand the information that needs to be in
the form and the degree of clarity required.  Dr. Fisher asked if EPA
agreed with the AHETF’s decision to destroy the records of women who
have a positive pregnancy test.  Mr. Carley explained that EPA expects
these women to be included in counts of interview participants, but
agreed with the decision to destroy their information.  Dr. Chadwick
noted that testing for literacy was not a requirement for participation
in this protocol.  Dr. Sean Philpott stated that the ICF as written is
too complex for the target population and should be simplified.

Dr. Fisher summarized that the protocol meets the requirements of 40 CFR
part 26, subparts K and L with some changes and recommendations.  The
AHETF should ensure that its test for understandability is objective and
should train interviewers in assessing understanding.  All forms need to
list Spanish speakers who can be contacted by potential subjects. 
Destroying the records of pregnant subjects is acceptable.  Regarding
photographs, any exculpatory language in the ICF should be removed.  Use
of films should be described in the ICF and restricted in their use
(i.e., used only for research or training purposes).  Task Force
observers should strive to avoid taking pictures of participants’
faces or should work to obscure their identity.  The Task Force did not
provide data on gender distribution and the expected ethnic distribution
(90 percent White and 10 percent Hispanic) was surprising; the AHETF
should describe the reasons for this distribution.  The reference to use
of interpreters and witnesses should be modified.  The phrase
“reasonable and appropriate” should be removed from the provision
for medical treatment.  Inconsistencies between the protocol and the SOP
should be resolved; most of this information can be described in the SOP
and referred to in the protocol.  Clarification of adverse event
reporting to the appropriate IRB is needed.  The recruitment poster
should list a Spanish-speaking contact and a local contact. 
Clarification of when consent is obtained is needed.

Dr. Prentice inquired about training of informed consent interviewers. 
Dr. Fish responded that no information was provided concerning how the
consent interview should proceed.  The documents mention standard
Collaborative IRB Training Initiative or National Institutes of Health
(NIH) procedures.  Dr. Prentice said that the Task Force should consider
including culturally relevant training.  Dr. Carriquiry remarked that
the Spanish version of the documents is inaccurate; the Spanish version
does not state that participation is voluntary until quite late in the
document.  Clarification that participation is voluntary is needed.  In
response to suggestions to obtain a back translation of the documents,
Dr. Fish explained that many interpreters find this inadequate and
prefer use of a certified translator to ensure that the language is
appropriate and culturally suitable.

Review of April 9-10, 2008 HSRB Meeting Report

Review Process

Dr. Lewis stated that the Board would review for subsequent approval of
its report for the April 9-10, 2008 HSRB meeting.

Public Comments

	Dr. Lewis invited oral public comment on the draft April 9-10, 2008
HSRB meeting report.  No oral public comments were presented.

Board Discussion and Decision on Report

Dr. Fisher opened discussion on the draft report by asking if Board
members had any suggested changes to the Board Recommendations on Review
and Format on the AEATF and AHETF Protocols (overall recommendations,
format of the protocols for subsequent HSRB review, format of Agency
presentations, and Task Force comments at the meeting) (p. 15, lines
25-45 and p. 16, lines 1-38).  No comments or changes were made.

Regarding the HSRB Consensus and Rationale related to scientific
considerations for the AEATF-II study protocols (p. 21, lines 42-46 and
p. 22, lines 1-36), no comments were made.  No comments were made for
the HSRB Consensus and Rationale related to ethical considerations (p.
26, lines 43-46 and p. 27, lines 1-2).

No comments were made for the HSRB Consensus and Rationale related to
scientific considerations for the protocol ICR A382 (p. 28, lines
30-36), studying the efficacy of two formulations of picaridin for
repelling stable flies.  The Board also had no further comments on the
HSRB Consensus and Rationale related to ethical considerations (p. 30,
lines 15-19).

The Board had no further comments on the HSRB Consensus and Rationale
related to scientific considerations for Carroll-Loye Biological
Research Completed Studies SCI-001.4 and SCI-001.5 (p. 32, lines 4-9),
testing LipoDEET 302 and Coulston’s Duranon for repelling mosquitoes
in two field environments.  No comments were made regarding the HSRB
Consensus and Rationale related to ethical considerations for these
protocols (p. 35, lines 10-13).

The Board did not make changes to the section of the draft report
describing the Board’s decision regarding future review of protocols
with planned deviations from prior IRB review (p. 35, lines 15-46 and p.
36, lines 1-2).  

Dr. Krishnan inquired if the overview on page 32 should be moved before
the science review in the report.  Dr. Philpott remarked that there is
some redundancy in the report regarding abstracts of the protocols, but
this will not occur in the future.  An abstract provided by EPA will be
included at the beginning of the review of a protocol, followed by the
science and ethics reviews.  The HSRB will review these abstracts to be
sure that issues the Board considers to be of importance are included. 
Dr. Fisher agreed to eliminate the subheading on page 32, line 20.

Dr. Fisher asked each Board member to approve the changes discussed
during this meeting.  All Board members in attendance at the conclusion
of this discussion approved the report.  Dr. Lewis added he will make
any necessary grammatical changes to the report.

Follow-up from Previous Day’s Discussion

	Dr. Fisher summarized Board conclusions for the next AHETF submission. 
The Board would like to see consideration of the recommendations it made
at this meeting.  In particular, an observer recording sheet describing
how worker behavior will be recorded should be included in information
provided to the Board, as should EPA’s analytic plan for its intended
use of the data.  It is difficult to determine if the 5x5 design will
provide useful data if the analytic plan is not included.  The plan can
be based on potential use as well as intended use.  EPA is welcome to
use consultants to develop its statistical approach.

EPA Review of Completed ICR Mosquito Repellent Efficacy Study A117

Background and Context

	Mr. Carley provided an overview of completed ICR mosquito repellent
efficacy study A117.  This study was a laboratory test of
already-registered repellent products for effectiveness against
mosquitoes of the genus Culex, a principal vector of West Nile Virus
(WNV).  The products were previously tested in the field against
mosquitoes of other genera.  Further testing was required by EPA to
support the proposed label claim to repel “mosquitoes which can vector
West Nile virus.”

	ICR protocol A117, submitted August 8, 2007, was reviewed favorably by
the HSRB in October 2007.  In early January 2008, ICR submitted a
revised protocol and ICF for informal EPA review, and met with EPA on
January 14, 2008, to discuss these documents.  On January 10, 2008,
the HSRB released the draft final report of its October 2007 meeting. 
In February 2008, ICR further revised the protocol and consent form,
obtaining final IRB approval on February 26, 2008.  Laboratory testing
was conducted on March 4, 2008, and the final report of testing was
submitted to EPA on April 9, 2008.

	Because specific events occurring in February and early March 2008 were
of specific concern for the ethics review, Mr. Carley described this
time period in further detail.  On February 8, 2008, ICR further revised
the protocol and ICF, reflecting EPA and HSRB comments.  On February 14,
2008, ICR submitted the revised protocol and ICF to Essex IRB, Inc.
(EIRB).  EIRB reviewed and approved the February 8, 2008 protocol and
reviewed and conditionally approved the February 8, 2008 ICF, with calls
for numerous changes on February 18, 2008; however, none of these
changes were of substantive significance.  Most were minor editorial
changes and a few were requests for insertion of standard language.  ICR
began recruiting potential subjects on February 18, 2008.  On February
19, 2008, EIRB notified ICR by email of the results of its February 8,
2008 review.  On February 20, 2008, ICR submitted a further revised ICF
to EIRB, which EIRB approved on February 25, 2008.  EIRB notified ICR of
the approval on February 26, 2008.

	On April 9, 2008, ICR submitted its final report and supplement for
protocol A117 to EPA.  On May 19, 2008, EPA emailed a request for
clarification on three points:  the rationale for the change in dosing
regimen reported as a deviation from the protocol, the number of
subjects, and the sequence and timing of recruiting actions.  ToXcel
responded to EPA’s request for clarification on May 21, 2008.  EPA’s
science and ethics reviews were completed on May 22, 2008.  On June 6,
2008, toXcel responded to EPA’s reviews.

EPA Science Assessment of ICR Study A-117:  Laboratory Repellency of Two
Registered Picaridin Products to Culex Mosquitoes

	Mr. Mark Suarez (OPP, EPA) provided EPA’s science assessment of ICR
study A117.  The test materials were two registered repellents
containing 10 percent picaridin:  Avon Skin-So-Soft SSS Bug Guard Plus
Picaridin Insect Repellent (Lotion) (Reg No 806-29) and Avon
Skin-So-Soft SSS Bug Guard Plus Picaridin Insect Repellent (Spray) (Reg
No 806-31).  The study objectives were to determine the efficacy of the
repellents to Culex mosquitoes in the laboratory to support proposed
label claims for efficacy against “mosquitoes which can vector West
Nile virus,” based on the hypothesis that the repellent samples will
provide 8 hours of protection from Culex quinquefasciatus mosquitoes in
the laboratory.  Because the objectives of the study did not include
establishing a typical consumer dose, the dose used was the standard
1 gram per 600 square centimeters, for consistency with the dose used
in earlier testing of the same repellents.

	Based on the Rutledge and Gupta (1999) meta-analysis, a target sample
size of 10 was needed to establish a complete protection time (CPT) of 8
hours ± 2 hours with 95 percent confidence.  A total of 12 treated
subjects were planned, to ensure an N of 10 even with as many as two
drop-outs.  One untreated control subject was chosen by lot.  All 12
treated subjects completed testing; all results were included in the
analysis.

	The criticisms of the scientific design of this protocol in EPA’s
review of September 24, 2007, and in the HSRB’s final report of the
October 2007 meeting were satisfactorily addressed through the revisions
made to the protocol before it was executed.  These criticisms include
the following:

use of 200 mosquitoes per cage

expansion of the statistical analysis plan to include descriptions of
how to assess normality of distribution, how to analyze non-normal data,
and statistics for all contingencies

consideration of the Kaplan-Meier test in place of the Q test

designation of an endpoint as “CPT”

addition of a form for attractiveness data

appending the product labels to the protocol

clarification of the reference to “dose range finding”

extension of the test duration to 10 hours.

	Final revisions to the protocol were made on February 8, 2008.  The
revised protocol was approved by EIRB on February 18, 2008, and executed
on March 4, 2008.  The execution was generally consistent with the
protocol, with one reported deviation.  Instead of treating subjects in
six pairs, subjects were treated in two groups of six.  This deviation
did not affect the experimental outcomes, but should have been done by
protocol amendment rather than as a deviation.

	The cages used in this study were 2x2x2 foot cubes, sitting on a table.
 The floor of the cage is mirrored to enable observers to see mosquitoes
on the underside of the subjects’ arms.  The sides and top of the cage
are screen wire.  On two of the sides there are two fabric sleeves that
are secured around the upper forearms of the subjects with large rubber
bands.  Because there are two sets of sleeves, it was possible to test 2
subjects at once, each testing a different repellent on each forearm.

	Two hundred mosquitoes were released into each cage.  The mosquitoes
were laboratory-reared, disease-free adult (3- to 8–day-old) females
that had been fasted for 24 hours and had never received a blood meal. 
Before subjects were treated with repellents, each subject established
attractiveness to the test mosquitoes by placing an untreated arm into
the mosquito-filled cage.  Once each subject had shown attractiveness to
mosquitoes, they were treated with one test repellent on each forearm
and testing began.  At 30-minute intervals, the untreated control
subject confirmed that the mosquitoes were still aggressive by putting
an untreated forearm into each cage.  If fewer than 5 mosquitoes landed
on the control subject’s arm within one minute, (as happened early in
the test), 200 fresh mosquitoes were added to each cage.  After the
aggressiveness of the mosquitoes was confirmed, treated subjects placed
their arms through the sleeves into the cage for 5 minutes.  This cycle
of confirmation of aggressiveness followed by 5-minute test exposures of
all treated subjects was repeated for 10 hours, or until a subject
experienced failure of efficacy on both arms, defined by a confirmed
bite on each treated arm.

	A failure of efficacy was defined as a “confirmed bite,” which was
one bite followed by another on the same arm within the same or the
subsequent 5-minute exposure period.  For test substance A, there were a
total of five bites (three unconfirmed and one confirmed).  For test
substance B, there were a total of six bites (two unconfirmed and two
confirmed).  CPT was defined as the time from treatment with the
repellent until the first confirmed bite, i.e., until the time of the
first bite confirmed by another (confirming) bite within the same or the
subsequent 5-minute exposure period.  CPTs were 9.417 ± 1.094 hours for
test substance A and 9.667 ± 0.421 hours for test substance B.  A
larger error was generated for substance A (lotion) because of an early
failure.

	EPA has concluded that the study provides scientific results that meet
EPA guideline standards, based on the study protocol as amended in
accordance with EPA and HSRB comments before testing began.  Both
products provided a CPT of approximately 9.5 hours against the WNV
vector Culex quinquefasciatus.

EPA Ethics Assessment of ICR Study A-117:  Laboratory Repellency of Two
Registered Picaridin Products to Culex Mosquitoes

	Mr. Carley provided EPA’s ethics assessment of protocol A117. 
Documents considered in this review were the primary report of ICR Study
A117 (MRID 47397701) and supplemental information to address
requirements of 40 CFR (MRID 47413601).  Additional documents considered
were EPA’s protocol review of September 24, 2007, toXcel’s response
to the EPA protocol review (October 17, 2007), toXcel summary of a
January 14, 2008 meeting with EPA, the HSRB’s final report of its
October 2007 meeting (January 17, 2008), toXcel’s response to EPA’s
request for clarification (May 21, 2008), and toXcel’s further
clarification of recruiting events (received June 6, 2008).  The last
communication was not received until after this ethics review was
completed.

	Three deficiencies were noted.  The IRB minutes did not explain the
basis for requiring changes in the ICF.  The IRB roster did not indicate
members’ experience sufficient to describe their chief anticipated
contributions to deliberations and EIRB procedures were not submitted. 
These gaps did not compromise EPA’s review.

	All comments from EPA and the HSRB were addressed in the revised
protocol and ICF of February 8, 2008.  These included the following:

revision of the discussions of risks and benefits in the protocol and
ICF

correction of the description of risk from repellents

addition of a discussion of environmental risks

discussion of risk of disease included in the ICF

addition of a description of attractiveness testing to the ICF

removal of the assertion that subjects are representative

justification or removal of the age cap at 55 years

adjusting ICF eligibility criteria to track with the protocol

clarification of descriptions of recruitment and consent processes

deletion of subject signature from data collection forms

harmonization of different discussions of control selection

clarification of how landing mosquitoes will be removed

	One protocol deviation was reported.  The protocol called for treating
subjects in six pairs; instead subjects were treated in two groups of
six.  The rationale provided was to simplify moving large numbers of
subjects in and out of the testing room.  EPA believes that this change
could have been anticipated and should have been made by amendment, with
prior IRB approval.  An unreported deviation involving initiation of
subject recruitment before final IRB approval of the ICF also occurred.

	On February 18, 2008, EIRB approved the February 8, 2008 protocol and
conditionally approved the ICF.  On February 18, 2008, the principal
investigator communicated with EIRB and was told that recruitment could
begin.  On February 18, 2008, ICR began calling previous subjects to
determine their interest and availability for the March 4, 2008 study. 
On February 19, 2008, EIRB notified ICR by email of its decisions made
on February 18, 2008.  On February 20, 2008, ICR revised the ICF and
submitted it to EIRB.  EIRB approved the revised ICF on February 25,
2008 and notified ICR of its approval on February 26, 2008.  On
February 26, 2008, ICR mailed copies of the approved ICF to candidates
who had expressed interest in participating in the study.  The
communication with EIRB on February 18, 2008, and calling of subjects
occurring on the same date were reported in the June 6, 2008 report
from toXcel.

	Both EPA and the HSRB expressed concern about the recruiting process in
their reviews of ICR protocol A117, and called for its clarification. 
The protocol description of the recruiting process was revised after the
October 2007 HSRB meeting to reflect EPA and HSRB comments.  The revised
protocol was approved by EIRB on February 18, 2008.  The recruiting
process initiated by ICR on February 18, 2008, and described in the
toXcel submissions dated May 22, 2008 and June 6, 2008, differed in
numerous respects from the process described in the revised protocol.

	Based on the toXcel submission of June 6, 2008, EIRB reportedly told
ICR on February 18, 2008 that “recruitment of study subjects could be
initiated since the protocol had been approved.”  However, no
documentation of the reported communication between the principal
investigator and EIRB personnel on February 18, 2008, has been provided.
 The June 3, 2008 email from EIRB on this matter does not mention any
prior communication on this subject.  Neither ICR nor toXcel has
reported, nor is there evidence of, any imminent hazard to subjects that
could have justified a change to approved research without prior IRB
approval.

	Based on its review, EPA has concluded that subjects were recruited
through a process differing from that prescribed in the protocol,
although this deviation from the protocol did not put subjects at
increased risk.  The subjects saw and signed the ICF only after final
IRB approval; therefore, the deviations from the protocol in the
recruitment process are unlikely to have seriously compromised the
informed and voluntary consent of the subjects, all of whom were
experienced subjects in ICR repellent tests.

	The applicable standards for this research are 40 CFR §26.1703,
prohibiting intentional exposure of pregnant or nursing women, or of
children under 18; 40 CFR §26.1705, requiring evidence of substantial
compliance with 40 CFR part 26, subparts A-L; 40 CFR §26.1303,
requiring documentation of ethical conduct; and FIFRA §12(a)(2)(P),
requiring fully informed, freely voluntary participation.  The
requirements of 40 CFR §26.1125 and §26.1601 for prior review of the
protocol by EPA and the HSRB were met.  The requirements of 40 CFR
§26.1303 and 40 CFR §26.1703 were met.

	The evidence that investigators initiated recruitment before final EIRB
approval of the consent form suggests that the conduct of the study may
not have been in substantial compliance with 40 CFR §26.1705 part 26,
subparts A-L.  The requirement of FIFRA §12(a)(2)(P) for fully
informed, fully voluntary consent of subjects may have been compromised
by initiation of recruitment before obtaining final IRB approval of the
ICF.

	EPA interprets 40 CFR §26.1111(a)(4) and (5) to mean that research is
not approved until both §26.1116 and §26.1117 have been satisfied;
however, EPA has learned that OHRP is considering guidance that would
allow recruiting to begin when the protocol and ICF have received at
least conditional IRB approval, as was true in this case.  The
initiation of this research before full IRB approval did not endanger
subjects, but led to procedural deviations from the protocol which may
or may not rise to the level of substantial non-compliance with 40 CFR
part 26, subparts A-L.

ICR A117 Charge Questions:

a.	Is this study sufficiently sound, from a scientific perspective, to
be used to assess the repellent efficacy of the formulations tested
against mosquitoes of the genus Culex?

b.   Does available information support a determination that this study
was conducted in substantial compliance with subparts K and L of EPA
regulations at 40 CFR part 26? 

	Clarifying Questions

	Dr. Philpott noted that the supporting information on ethical conduct
lists a third product to be tested (towelettes) and asked if this was a
mistake.  Mr. Carley responded that the protocol calls for testing of
two products and two products were tested.  Dr. Philpott inquired if EPA
has requested and received a telephone script for the recruiting
process.  Mr. Carley replied that there is a passage in the protocol
characterized as a “script” but it is vague and lacks detail.  EPA
does not know the exact context or content of the recruiting script.

	Dr. Menikoff questioned how the change from six pairs of two subjects
to two groups of six subjects affected the approval of the protocol. 
Mr. Carley explained that the change was reasonable, given that it
simplified the testing and had no impact on results or ethics.  The
investigators should have submitted a protocol amendment to make this
change because it was possible to anticipate and was not made in
response to a hazardous testing situation. 

	Dr. Menikoff requested a description of the IRB communication with the
principal investigator on February 18, 2008.  Mr. Carley explained that
on June 6, 2008, ICR included a copy of the June 3, 2008 email from IRB
to the principal investigator as supporting documentation that
recruiting could be initiated before receipt of the final approved ICF. 
The letter does not specifically mention this situation, in which an
unconditional approval of the protocol and conditional approval of the
ICF had been received.  Dr. Menikoff noted that EIRB did not
specifically claim that the events violated its terms and said that it
appeared that the first telephone call to potential subjects may have
been to check availability and the protocol may not have been mentioned.
 The call might have been supplemental to what was described in the
protocol as the first recruitment call.  Mr. Carley commented that EPA
does not know the exact content of the call.  The protocol indicates
that an initial contact will be made to ask subjects if they are
interested and available for the study, and then subjects will be
informed about the study and the ICF will be described.  If subjects
agree to participate, the ICF will be sent to them and after receipt is
confirmed, ICR calls the subject and reviews the ICF within the next few
days.  There appears to be insufficient time for this process because of
when the approved ICF was sent.  A compression of the protocol (fewer
steps and fewer contacts with potential subjects) may have occurred. 
Dr. Menikoff noted that the subjects other than the three who did not
receive the ICF were offered a chance to discuss the ICF.  The fact that
three people did not receive the ICF appears to be a violation.  Mr.
Carley stated that there were procedural violations, but the substantive
steps of the process were in place.

	Dr. Prentice commented that ICR began recruiting on February 18, 2008,
before final approval of the ICF and inquired about the nature of the
revisions to the ICF and whether they were substantive.  Mr. Carley
explained that there were approximately 15 changes to the ICF, and all
but two or three were minor editorial changes.  The exceptions were the
request by EIRB to insert standard language, including a paragraph that
informs subjects about resources on participation in studies.  These
resources include only links to the main portals to OHRP and the NIH,
and seem unlikely to be helpful concerning participation in this type of
study.  EPA believes that addition of this material is unnecessary and
will contact EIRB about this matter.  Dr. Prentice asked if there was
any documentation of the communication between the principal
investigator and EIRB that occurred on February 18, 2008.  Mr. Carley
responded that EPA had received only the June 6, 2008 clarification that
the ICR staff was “told” they could proceed with recruiting.  This
email from EIRB is dated June 3, 2008, and does not refer to the prior
communication.  Mr. Carley added that he had not had a chance to speak
with ICR about this issue in detail.

	Dr. Prentice remarked that the recruiting calls on February 18, 2008
could indicate use of an unapproved script, but the HSRB lacks the
information needed to decide if this is the case.  Mr. Carley agreed
that ICR’s characterization of their recruiting process was highly
inadequate.  Dr. Krishnan inquired if checking availability is
considered part of the recruiting process.  Mr. Carley replied that
this is part of the process.  Dr. Krishnan noted that ICR’s reply
states that potential subjects signed the ICFs, but does not say when
they signed them.  Mr. Carley explained that all subjects signed the ICF
on March 4, 2008, the day of testing.  There was insufficient time for
the subjects to sign the forms and mail them back to ICR.  A statement
in the supplemental materials states that all subjects signed the
approved ICF on the morning of March 4, 2008.  Dr. Krishnan noted that
in ICR’s response, they indicate that all subjects agreed to travel to
ICR to have their questions answered, and agreed to be prepared to sign
the ICF on the day of testing.  Mr. Carley clarified that all subjects
were invited to travel to ICR to discuss the ICF but none did.  All
subjects arrived on the morning of testing and discussed the ICF at that
time.  All the subjects participating in the protocol were experienced
subjects who had participated in previous ICR protocols.

	Dr. Parkin inquired when EIRB procedures had been previously submitted
to EPA.  Mr. Carley indicated that EIRB had submitted its procedures to
EPA approximately 1 year ago.  Dr. Parkin stated that because EPA had
received no documentation of EIRB procedures, the Board has no knowledge
concerning whether EIRB was following its procedures in review of this
protocol.

Public Comments

	Dr. Robin G. Todd of ICR, Mr. Niketas Spero of ICR, Dr. Ralph Piedmont
of Loyola College, and Mr. Andrew Pechko of Avon Products, Inc., on
behalf of ICR

	Mr. Niketas Spero (ICR) explained that on February 18, 2008, ICR
received a telephone call from EIRB, as is typical when they complete a
review.  EIRB explained that the protocol was approved and the ICF was
conditionally approved.  During the call, Ms. Karen Radcliffe (EIRB)
was asked if recruiting could begin.  She responded that recruiting
could begin, but the ICF could not be discussed.  Later that day, Mr.
Spero began the recruitment process but limited the substance of the
telephone calls to potential subjects to asking about interest and
availability.  March 4, 2008 was chosen as the test date to allow
sufficient time to raise the mosquitoes and recruit a sufficient number
of subjects.  Only subject availability was discussed on the call.  When
the approved ICF was received on February 26, 2008, Mr. Spero called the
available subjects back and discussed the ICF and protocol.  At this
time, subjects also were offered the opportunity to come to ICR and
discuss the ICF in further detail, but all declined this opportunity.

	Regarding the deviation of the protocol from six groups of two to two
groups of six, this was not considered until the planning meeting on the
day before testing.  ICR decided it would be easier to test 12 people if
there were two dedicated times for exposure in the cage.  There was
insufficient time to amend the protocol.  Mr. Spero acknowledged that
the investigators should have considered this approach earlier.

	Dr. Fisher inquired why the test date of March 4, 2008, had been
chosen.  Mr. Spero explained that this was based on the sponsor’s need
to present its report to EPA.  Also, raising mosquitoes needs to begin
approximately 2 weeks in advance, and ICR believed all paperwork would
be approved by this date.  Dr. Fisher asked when ICR had notified EIRB
of the change in the number of test groups.  Mr. Spero stated that this
was noted as a deviation and EIRB was contacted 10 days later.  Dr.
Fisher requested clarification regarding testing of the towelette
product.  Mr. Spero explained that ICR evaluated only the aerosol and
pump spray.  Testing will support the towelette product because the
three AIs are the same.  Although only two product types were tested,
the regulatory requirements for the towelette product also will be met. 
Mr. Andrew Pechko (Avon Products, Inc.), on behalf of ICR, explained
that the DASH30 towelette uses the same liquid formulation that is in
the pump spray, so this efficacy test supports the towelette product.

	Dr. Fisher asked if ICR had provided its written recruitment script. 
Mr. Spero explained that ICR did not use a script for this study, but
has one for future use that has been approved by EIRB.  EIRB approved
the vague script described in the protocol.  During the first call to
subjects (on February 18, 2008), subjects were informed that testing
would occur on March 4, 2008, and asked if they were interested and
available.  If they were, subjects were told that the ICF could not be
discussed because it had not yet been approved, but that once it was
approved, they would be contacted to discuss the form.

	Dr. Krishnan inquired if all subjects had seen the ICF only the day
before testing, or if they saw it on an earlier date.  Mr. Spero
explained that the ICF was approved on February 26, 2008.  Subjects
were called on that day and the ICF was discussed (in particular, the
exclusion criteria and parameters of the study), and it was mailed to
subjects the same day.  Subjects were called again on February 29, 2008,
after they had received the ICF.  This second call lasted approximately
5 to 10 minutes and indicated that all participants read the ICF and
were comfortable with ICR procedures.

	Dr. Philpott asked if subjects received compensation for traveling to
ICR to discuss the ICF, if they chose to do so.  Mr. Spero responded
that there was no compensation for this activity.  All subjects declined
to come in to discuss the ICF because they had participated in many
previous studies.  Dr. Philpott inquired about the subjects who did not
receive the ICF by mail and whether they had an opportunity to review
the ICF before testing began.  Mr. Spero explained that the ICF was read
over the phone (February 26, 2008) to those who did not receive it in
the mail.  The test day was scheduled to begin at 6:00 a.m., which
provided ample time for subjects to read the ICF and ask questions about
it before signing.

	Dr. Fisher asked if the boilerplate language that EIRB suggested
inserting included any language from the EPA Web site or just OHRP.  Mr.
Spero answered that it did not include EPA language.  Mr. Carley
clarified that there were three insertions of standard language, but no
major changes to the ICF.

Board Discussion

	Scientific Considerations – ICR A117

	Dr. Chambers opened the Board’s science discussion of ICR protocol
A117.  Use of a standard consumer dose of the repellent and counting
bites as opposed to landings was previously approved by the HSRB and
EPA.  She agreed with Mr. Carley that the change to two groups of six
participants did not affect the science of the protocol.  She concluded
that the results were scientifically valid.

	Dr. Krishnan agreed with EPA and Dr. Chambers that, given the purpose
of the study, the results are acceptable and sound.  He noted that there
was a typographical error in the table on page 4 referring to product
numbers.

	Dr. Young agreed that the results were scientifically valid, but she
commented further on issues related to mosquito studies in general.  She
expressed concern about conclusions drawn as a consequence of the study.
 The hypothesis is that the repellents will offer 8 hours of protection;
however, at 3 hours the product failed for one of the subjects; this
represents 8 percent of the subjects and is thus a significant finding. 
She expressed concern about how this would affect the labels of the
products.

	Dr. Young noted the heavy reliance on the Rutledge and Gupta (1999)
meta-analysis paper for the sample design.  She stated that the
designers of this protocol over-relied on the analysis and that the
meta-analysis itself is flawed.  This analysis is reasonable to use for
rough power calculations, but in the documentation of the protocol there
is an assumption that if there are the same number of subjects in the
experiment, the same conclusions can be drawn as those described in
Rutledge and Gupta.  This is untrue.  In addition, many of the articles
used in the meta-analysis are flawed and the meta-analysis is weighted
toward large studies with small means; traditional modeling of means and
variance were not used.

	Dr. Young also noted problems related to right-censoring of the data. 
Eleven subjects in this study had no bites at 10 hours, but one person
was bitten at 3 hours.  If these data are analyzed by Kaplan-Meier
analysis, the analysis falls apart when attempts are made to compute the
mean because more than 90 percent of the data points are censored.  This
is problematic for statistical analysis of the data.  To base the
conclusion of 8 hours ± 2 hours of protection on the Rutledge and Gupta
article is seriously flawed.  Careful consideration of how to analyze
these studies in the future is needed.  If the results claim to define a
mean, the mean is approximately 8 hours of protection; however, again
use of the Kaplan-Meier equation would be flawed.  Dr. Fisher asked Dr.
Young if she had any recommendations or suggestions for future studies. 
Dr. Young suggested that there are different models that would be more
suitable for this work, but nothing is available in the published
literature comparable to the Rutledge and Gupta paper for discussion of
study power.  Re-analysis of the studies included in the Rutledge and
Gupta analysis also might be helpful.

	Dr. Fisher noted that this issue continues to arise in Board
discussions, but the Board has been unable to recommend useful
alternatives.  Dr. Carriquiry explained that the main problem is the
very small sample size.  To resolve this issue, either a larger sample
size or longer observation period (perhaps 15 hours to observe product
failure) is needed.  Dr. Young stated that failure should reach at least
50 percent before the study stops.  Dr. Fisher inquired if parametric
statistics were appropriate for this sort of study.  She added that
increasing the observation period to 15 hours was probably unreasonable.
 Dr. Young remarked that, on average, stating that the product offers 8
hours of protection is reasonable, but this will not include those for
whom the product fails early.

	Dr. Lebowitz agreed that there was no method for analyzing the data in
a way that would be statistically significant.  The policy issue in
question is what sort of information to include on the product label,
given the evidence.  He agreed with Dr. Young’s frustrations, but
noted that while these discussions have arisen before, changes in sample
size or length of observation period have not occurred.  He stated that,
regarding the answer to the scientific charge question, he agreed with
Drs. Chambers, Krishnan, and EPA, that the data are sound.  How EPA uses
the data to inform labeling is a regulatory issue for which the Board is
not skilled to provide advice.

	Dr. Fisher remarked that the charge states that, from a scientific
perspective, in conjunction with other available information, the study
established what it hypothesized.  She asked if the Board wanted to
include a statement that the data are sufficient for estimating a range
of response times, but not an average.  Labeling decisions are beyond
the purview of the Board.

	Dr. Young stated that it was not possible to obtain an unbiased
estimate of the mean or derive the standard deviation using these data. 
Dr. Brimijoin stated that he agreed with the consensus that the study
has adequate scientific validity and answers the charge question.  He
agreed with Dr. Lebowitz that the science was as good as or better than
that in the typical insect repellent study.  The Board can offer any
suggestions it chooses to EPA to guide them on regulatory decisions, but
these are outside the purview of the Board.  He expressed sympathy with
the frustrated statisticians regarding the study design, power, and
interpretation.  The raw data from these studies are available and the
Board can revisit them to try to determine a more appropriate analytical
approach.  Dr. Brimijoin stated that these data are consistent with
current “best practices” for insect repellent studies.  Two changes
are needed to improve the studies.  Insect repellent investigators must
rethink their study designs in light of the proposed statistical
analyses.  The statistical community also should consider which
measurements and endpoints would be appropriate for these studies.  He
commented on survival analyses for treatments for fatal conditions in
which the endpoint is 100 percent death; however, such studies do not
need to achieve a 100 percent death rate for detection of the effects of
treatment to be possible.  Unlike most cancer chemotherapeutics, insect
repellents work well, approaching a 100-percent success rate over times
that are reasonable for testing.  It would not be practical to test
repellents in a research setting for much longer than 8 to 10 hours.  He
disagreed with Dr. Young’s assessment that because all data are
censored except for one data point that the calculated mean is
meaningless.  He noted that if 99 out of 100 subjects were protected for
more than 10 hours but one was protected for only 2 hours, the
protection range would not be 2 to 200 hours; nearly everyone in this
example received complete protection.  Dr. Brimijoin stated that he was
not prepared to offer EPA advice on regulatory issues or ways to treat
the data, but the data presented at this meeting are meaningful for
protection; a mix of common sense and mathematics should be applied to
assessment of the data.

	Dr. Carriquiry countered that, given the data from this study, she
would expect there to be 8 to 10 failures in a study of 100 subjects. 
If investigators wish to observe a rare event, they must increase their
sample size.  She stated that she would not oppose the Board’s
conclusion that the study is acceptable, but it was problematic that the
data did not allow estimation of an unbiased mean.  Dr. Brimijoin
commented that instead of increasing sample size, another option would
be to increase mosquito pressure to perhaps 10 times that used in these
experiments; another alternative would be to decrease the amount of
repellent used.  Dr. Young remarked that it would be possible to
estimate the proportion of users who would receive less than 8 hours of
protection if the sample size was larger; current data would estimate a
failure rate of 8 percent.  Dr. Fisher noted that the typical sample
size of 10 was acceptable in part to minimize risk; however, laboratory
experiments do not carry the same degree of risk as field studies.

	Dr. Lebowitz interjected that the Board was making assumptions about
what might or might not happen, given changes to the protocol.  The
Board can only carefully consider and discuss the information it is
given.  The Board also does not have any adequate recommendations for
changes for future designs.  He recommended that EPA consider the issues
raised at this meeting and request that the HSRB discuss them further,
if the Agency desires.  Dr. Chambers stated that EPA needs some
consistent guidelines to help the Board focus on scientific issues,
rather than study design and regulatory matters.  Dr. Fisher stated that
in her opinion, the data were useful for assessing efficacy, but not for
determining a mean or standard deviation; however, the study was
well-executed and questions about limiting the uses of the data have
arisen in the past.

	Dr. Ralph Piedmont (Loyola College), on behalf of ICR, stated that he
agreed that the Rutledge and Gupta study is inadequate and flawed, but
it represents the current state of knowledge regarding repellent
studies.  In designing the study, the goal was to be conservative; the
Rutledge and Gupta study was the only report available for guidance on
power.  With a sample size of 12, it is possible to have a mean estimate
of protection of up to 10 hours ± 2 hours.  Given the current state of
the science, these data indicate that the product works for 8 hours ±
2 hours.  He agreed that use of the actual mean is not supported because
it is biased and 95 percent confidence intervals cannot be justified;
however, the use of 12 subjects was scientifically justified given the
existing knowledge.  Dr. Lebowitz agreed that the Rutledge and Gupta
paper was insufficient, with the Board’s assessment of the statistics
of the study, and with the conclusion that the Board does not believe
the evidence is sufficient to agree with conclusions about means and
standard deviations drawn using the results of the study.

	Dr. Fisher summarized that the Board consensus was that the study was
conducted using best practices following an approved protocol; however,
the Board does not believe that the data can be used to develop a mean
and standard deviation.  EPA will decide if this is possible and how
they will use the data.  She also agreed that the investigators had used
current best practices to design their study.

	Ethical Considerations – ICR A117

	Before the discussion began, Dr. Fisher reminded the Board of its
consensus regarding protocol deviations:

Any study executed prior to IRB approval of the ICF and the protocol, or
changed in ways that were not approved by the IRB will be judged by the
Board as failing to meet the applicable requirements of §40 CFR part
26, subpart K. 

If EPA submits to the Board for review a completed protocol with
scientific deviations from the original protocol reviewed by the Board,
the EPA review of the completed protocol should provide the Board with
EPA's opinion regarding why the deviation did not meet the requirement
for re-review and why the protocol still meets the applicable
regulations.

	Dr. Philpott opened the ethics discussion of ICR protocol A117 by
stating that the study had been adequately designed to minimize risks. 
The relevant risks included participating in research conducted under
hot and humid conditions, adverse events associated with use of the
products, exposure to insect bites, and exposure to vector-born
illnesses.  The risks associated with the hot and humid research
environment are minimized by limiting participants to 5-minute intervals
in this environment.  The risks associated with use of the product are
minimized because the AI (picaridin) is registered and widely used,
there are adequate stopping rules, and people with skin reactions to the
products are excluded from the research.  Bites are expected, but risks
from exposure to bites are minimized by excluding participants with a
history of severe reactions to insect bites and by close monitoring and
providing medical care.  The risk of vector-borne disease is minimized
because the research uses laboratory-raised mosquitoes that have not had
a previous blood meal.  The study investigators also contacted
participants after the research was complete to ask about any adverse
events.

	Key issues of concern related to this protocol include the protocol
deviation and consent issues.  Protocol deviations that appear planned,
but are not reviewed by an IRB arise frequently.  In this case, the
change from testing six pairs of two subjects to testing two groups of
six subjects was a planned violation because the investigators discussed
this change the day before testing began.  It was possible for the
investigators to call EIRB and obtain some sort of approval for the
change, but they did not do so.  The Board has explicitly explained that
federal regulations state that the only allowable protocol violations
are those that occur to protect the subjects.  In addition, ICR did not
contact EIRB about this violation until 10 days after the research took
place, and did not include EIRB’s response in the documentation of the
research.

	Recruitment took place prior to revisions and final approval of the
ICF, although the Board has been told that ICR received verbal approval;
however, on page 92 of the report there is a written statement from the
Chair of the IRB stating that a study may not commence any further
research activity, including scheduling, until a stamped ICF is
received.  This statement is dated August 2007.  ICR has clearly
violated IRB rules regarding recruitment.  Because of this,
Dr. Philpott stated that it was his opinion that the protocol is not in
compliance with 40 CFR part 26 and he recommended that EPA cannot use
the results of this study.

	Dr. Menikoff stated that he disagreed, to some extent, with Dr.
Philpott’s conclusions.  He noted that he had objected to the
Board’s consensus on protocol violations previously, and believed
other Board members did as well, because the statement does not include
a consideration of intent.  A discussion of intent occurred, but
statements related to intent were eliminated from the consensus.  The
Board is asked to consider if a protocol is in “substantial”
compliance; the Board consensus statement appears to demand
“perfect” compliance.  Given Mr. Carley’s thorough review, the
Board may not be able to approve any research if it demands perfect
compliance.  Dr. Fish added that this particular study was conducted
before the Board created its statement and thus the consensus statement
should not be applied as rigorously.  Dr. Menikoff stated that in terms
of consent and recruitment, differences in facts exist.  The written IRB
materials differ from statements that the IRB emailed or said to the
investigators.  This matter can be investigated.  If EIRB informed the
investigators by telephone call that recruitment could begin despite the
lack of a finalized ICF, then the study is not in substantial
noncompliance.  Calling people who previously agreed to be contacted
about participation in these studies to ask if they are available on a
certain date should be acceptable.  Dr. Menikoff added that reading the
ICF over the telephone to those who did not receive the ICF in the mail
also is not necessarily a violation.

	Regarding the change in protocol design, this also may not rise to the
level of substantial noncompliance, given that it did not increase
subject risk; however, because the deviation was planned, Dr. Menikoff
stated he was more likely to agree with Dr. Philpott on this matter.

	Dr. Fisher agreed that if EIRB gave permission to begin scheduling
activities, the sponsor is in compliance.  The HSRB has questioned IRB
activities in the past; if this IRB did not behave in a legitimate
manner, the HSRB should make a strong statement to this effect.  Dr.
Menikoff agreed, but added that this may be complicated by apparent
changes in recruitment policies taking place through OHRP and the IRBs.

	Dr. Chadwick stated that the Board is continually disappointed by
investigators who do not follow their own protocols; however, it may be
unreasonable to expect 100-percent compliance.  Many investigators view
protocols as guidelines, while regulatory entities and the HSRB appear
to view them more as “written in stone.”  The Board needs to
recognize that two types of changes to protocols occur:  deviations,
which affect neither science nor risk, and violations, which affect
subject protection, or science, or both.  In this case, a deviation
occurred.  It is true that the investigators did not follow the written
protocol, but this did not result in subject harm or damages to
scientific validity, which the Board is charged with reviewing.  The
Board must carefully consider wording; the HSRB believes that
recruitment means enrollment, but a member of an IRB may believe that
recruitment includes asking about availability.  Dr. Chadwick cautioned
sponsors that other agencies have the policy of “if it is not written,
it did not happen.”  Documentation is important for regulatory
agencies, and justifying an action on verbal advice is a bad choice. 
Dr. Chadwick concluded that in his opinion, ICR protocol A117 is
ethically acceptable.

	Dr. Fish agreed with Dr. Chadwick.  Ethicists may consider recruitment
to be part of the informed consent process because most recruitment
processes provide information about the study to the potential subjects,
and it is important that this information is approved by an IRB.  In
this case, however, recruitment involved knowledgeable subjects and
information on this study was not provided to them, except to tell them
that the study would take place on March 4, 2008.  If this is the only
information provided and no specifics of the study were discussed before
approval of the ICF was obtained from EIRB, recruitment is in compliance
with the intent of the regulations.

	Dr. Prentice agreed with Drs. Chadwick, Fish, and Menikoff.  He noted
that the risk-benefit balance is highly important to the IRB’s
approval of a study.  In this case, the primary risk is exposure to
mosquito bites and the discomfort associated with such bites.  There is
no direct benefit to the subject, only to society for registration of an
alternative efficacious repellent against mosquitoes that harbor WNV. 
If the Board declines to accept the data, the risk-benefit balance has
been negated for every subject, which is a serious matter.  Dr. Prentice
agreed that any unethical or significant deviations should not be
ignored, but in this case, the deviations are minor.  The deviations
indicate that the system is working; a deviation occurs, is reported,
and a mitigation plan is put in place.  He disagreed with Dr.
Philpott’s opinion that the results should not be used by EPA.

	Dr. Philpott agreed with the points raised by his fellow Board members,
but expressed concern that, based on the written materials reviewed by
the Board, the protocol is not in compliance with ethics regulations. 
He was concerned that the Board might be setting a precedent by making
value judgments.  In light of the Board consensus on protocol
deviations, Dr. Philpott concluded that he could not recommend EPA use
of the data.

	Dr. Lebowitz noted that for the National Incident Management System,
written or oral commands are acceptable, except in unusual cases where
the regulations require a written command.  He agreed with Dr. Chadwick
regarding the difference between deviations and violations.  If the
Board can conclude that neither the scientific nor ethical standards
were violated, deviations should be accepted as “human error.”  The
Board should not be so rigid that it cannot operate within this system.

	Dr. Prentice noted that slide 32 of EPA’s review stated “led to
procedural deviations from the protocol which may be deemed to rise to
the level of substantial non-compliance with subparts A through L of 40
CFR part 26,” but the charge asks if the “available information
supports a determination that this study was conducted in substantial
compliance with subparts K and L of EPA regulations at 40 CFR part
26.”  “Substantial compliance” indicates that a level of
compliance less than 100 percent would be acceptable.  This protocol was
conducted in sufficient compliance that the Board does not have concerns
about the science or blatant violations of the regulations.  Dr.
Prentice stated that the language in the charge question is limiting and
should be revisited by EPA.

	Mr. Jordan explained that the charge question language mirrors the
standard language in EPA’s regulations.  EPA may rely on a completed
study if there is evidence to indicate it substantially comports with
the regulations in subpart L.  EPA does not require 100-percent
compliance; the drafters of the regulations (Mr. Carley and Mr. Jordan)
wanted to include the notion of acceptability.  EPA recognizes that
demanding 100-percent compliance would mean that a single deviation
would result in rejection of all protocols and would result in large
numbers of repeated studies; EPA does not intend for such a situation to
arise.  The regulations ask the Board to make value judgments. 
Concerning the Board consensus on protocol deviations, Mr. Jordan
agreed with the first point that an investigator who does not follow the
approved protocol fails to meet subpart K of the regulations.  But
subpart Q, which asks if the study is acceptable for EPA, indicates that
“substantial compliance” will suffice.  The deviations EPA notes are
deviations that investigators should not make knowingly, but EPA
recognizes that such deviations do happen and when they do, they should
be reported and corrected.  Mr. Jordan stated the HSRB’s statement was
acceptable if it does not affect the “substantial compliance”
standard of subpart Q.

	Dr. Philpott commented that according to the EIRB written statement,
recruitment could not be started until the approved ICF was received. 
Mr. Jordan explained that the August 2, 2007 approval was part of the
IRB review cycle preceding Board review in October 2007.  Before the
information came to the Board, a final protocol with approval of the IRB
was received.  The protocol was then revised and the telephone calls
between the sponsor and EIRB occurred in a different cycle of IRB
review.  Dr. Philpott noted that the language in the letter was
“boilerplate” and that the HSRB has not had access to IRB procedural
language because it is Confidential Business Information.  The Board
cannot assume that EIRB changed its language in the absence of written
information that indicates that beginning the recruitment process before
receipt of the approved ICF was acceptable.  Dr. Philpott stated that he
was reluctant to approve this study, and strongly recommended that no
study should be commenced in any way without final approval and a
stamped ICF.

	Dr. Fisher interjected that the Board must consider two issues that
relate to recruitment and relate to the change in study design.  She
agreed that the changes did not result in increased risk to
participants; the Board should consider whether this constitutes
“substantial” noncompliance.  The Board must take at face value the
claim from the sponsor (and the email dated June 3, 2008) that ICR was
allowed to begin recruitment before receipt of the finalized ICF.  The
Board may disagree with EIRB’s decision, but given that the decision
may be based on new directives from OHRP, this decision may not be a
violation of EIRB procedures.

	Dr. Philpott acknowledged that his was the minority opinion.  He would
accept the consensus to approve the study, but wanted to include a
strong statement that written confirmation of changes is needed and
verbal information is unacceptable.  Dr. Menikoff agreed that EPA should
verify changes in EIRB policies.  Dr. Fisher noted that sponsors should
not accept IRB procedures that may place them in jeopardy of having a
study disapproved.  

	Dr. Fisher remarked that the change in the study design was a deviation
that did not increase risk to subjects, but the fact that ICR waited 10
days to report this deviation to EIRB is troubling.  The sponsors should
have immediately emailed EIRB about this change.  The study was in
substantial compliance, but in the future the Board may judge more
harshly these non-documented changes.  Dr. Philpott added that the Board
must explicitly state that it will not accept hearsay as evidence, but
instead will require clear written documentation of changes and
permissions; the investigators and EPA must ensure that such information
is made available to the Board.  Dr. Chadwick countered that a rule
saying in effect, “If it is not written, it does not exist,” would
be too stringent.  Dr. Brimijoin agreed.

	Dr. Fisher stated that the majority of the Board agreed that the
protocol was in substantial compliance with the relevant regulations. 
The Board also should detail the non-substantial, but important
deviations that occurred.  Drs. Chadwick, Fish, Menikoff, and Philpott
will draft this statement.  

	Dr. Fisher recommended changing the Board’s consensus statement on
protocol deviations.  The Board always has the option of recommending
rejection or acceptance of a protocol.  The language of this statement
needs to be clarified, especially with respect to the issue of
“substantial” compliance.  Dr. Chadwick agreed with Mr. Jordan that
the deviations in this protocol represented violations of the
regulation, but it is more difficult to decide if the protocol is in
substantial non-compliance.  The Board’s statement should incorporate
subpart Q; compliance, or lack thereof, with subpart K may be clear, but
the Board will need to consider more thoroughly if a protocol is not in
compliance with subpart Q.  Dr. Menikoff stated that he did not object
to such changes, but was not convinced it would greatly influence Board
review of protocols.  The Board is rarely asked to judge if a protocol
is in 100-percent compliance.  Dr. Fisher stated that the Board’s
report for this meeting would include amendment of the Board’s
statement on protocol deviations, incorporating consideration of
compliance with subpart Q.

Concluding Remarks

	Mr. Jordan provided a preview of the October 2008 HSRB Meeting agenda. 
The Board will review at least three completed insect repellent studies
conducted in accord with previously reviewed protocols.  EPA also is
analyzing literature on potassium triiodide, involving intentional
dosing in human subjects to evaluate toxicities.  EPA may wish to use
this information for risk assessment activities, and if so, the Board
will review these studies.  The Board also will review additional
protocols from the AEATF and AHETF.  AEATF will present a scenario to
measure exposure from aerosol products.  AHETF is currently choosing
protocols to present for review.

	Dr. Fisher asked if there is existing pre-rule data that the Board must
review in advance of evaluation of open-cab scenarios planned by AHETF. 
Mr. Jordan explained that because these studies were initiated before
the rule was in place and did not report a toxic endpoint, Board review
is not needed.  However, EPA will work to provide the Board with data
and conclusions of this data in a summary form.

	Dr. Lewis thanked Board members for their efforts and his EPA
colleagues for their preparation and presentations given at this
meeting.  He stated that the next HSRB meeting will take place October
21-24, 2008, and will be announced in the Federal Register.  The Board
will release its April 2008 meeting report, prepare its report for this
meeting and send drafts to Drs. Fisher and Lewis within the next 2
weeks.  Dr. Lewis adjourned the meeting.

Respectfully submitted:

Paul I. Lewis, Ph.D.

Designated Federal Officer

Human Studies Review Board

United States Environmental Protection Agency

Certified to be true by:

Celia Fisher, Ph.D.

Chair

Human Studies Review Board

United States Environmental Protection Agency

NOTE AND DISCLAIMER:  The minutes of this public meeting reflect diverse
ideas and suggestions offered by Board members during the course of
deliberations within the meeting.  Such ideas, suggestions, and
deliberations do not necessarily reflect definitive consensus advice
from the Board members.  The reader is cautioned to not rely on the
minutes to represent final, approved, consensus advice and
recommendations offered to the Agency.  Such advice and recommendations
may be found in the final report prepared and transmitted to the EPA
Science Advisor following the public meeting.

Attachments 

Attachment A 		List of HSRB Members

Attachment B 		Federal Register Notice Announcing Meeting 

Attachment C 		Meeting Agenda 

 

Attachment A

EPA HUMAN STUDIES REVIEW BOARD MEMBERS

Chair

Celia B. Fisher, Ph.D.

Marie Ward Doty Professor of Psychology

Director, Center for Ethics Education

Fordham University

Bronx, NY

Vice Chair

William S. Brimijoin, Ph.D. **

Chair and Professor 

Molecular Pharmacology and Experimental Therapeutics

Mayo Foundation

Rochester, MN

Members

Alicia Carriquiry, Ph.D.

Professor 

Department of Statistics

Iowa State University

Ames, IA

Gary L. Chadwick, PharmD, MPH, CIP

Associate Provost

Director, Office for Human Subjects Protection

University of Rochester

Rochester, NY

Janice Chambers, Ph.D., DABT

William L. Giles Distinguished Professor

Director, Center for Environmental Health Sciences

College of Veterinary Medicine

Mississippi State University

Mississippi State, MS

Richard Fenske, Ph.D., MPH ** 

Professor

Department of Environmental and Occupational Health Sciences

University of Washington

Seattle, WA

Susan S. Fish, PharmD, MPH

Professor, Biostatistics & Epidemiology

Boston University School of Public Health

Co-Director, MA in Clinical Investigation

Boston University School of Medicine

Boston, MA

Suzanne C. Fitzpatrick, Ph.D., DABT 

Senior Science Policy Analyst

Office of the Commissioner

Office of Science and Health Coordination

U.S. Food and Drug Administration

Rockville, MD

Dallas E. Johnson, Ph.D.

Professor Emeritus

Department of Statistics

Kansas State University

Manhattan, KS

Kannan Krishnan, Ph.D.

Professor

Département de santé environnementale et santé au travail

Faculté de médicine 

Université de Montréal

Montréal, QC, Canada

Michael D. Lebowitz, Ph.D., FCCP

Retired Professor of Public Health (Epidemiology) and Medicine

Research Professor of Medicine

University of Arizona

Tucson, AZ

Lois D. Lehman-Mckeeman, Ph.D. *

Distinguished Research Fellow, Discovery Toxicology

Bristol-Myers Squibb Company

Princeton, NJ

Jerry A. Menikoff, M.D.  

Director, Office of Human Subjects Research

Office of the Director

National Institutes of Health

Bethesda, MD

Rebecca Parkin Ph.D., MPH 

Associate Dean for Research and Public Health Practice

School of Public Health and Health Services 

The George Washington University

Washington, DC

Sean Philpott, Ph.D., M.Bioethics 

Science and Ethics Director

Global Campaign for Microbicides 

Program for Appropriate Technology in Health

Washington, DC 

Ernest D. Prentice, Ph.D. 

Associate Vice Chancellor for Academic Affairs 

Professor of Genetics, Cell Biology and Anatomy

Professor of Preventive and Societal Medicine

University of Nebraska Medical Center

Omaha, NE

Richard R. Sharp, Ph.D. *

Director of Bioethics Research

Department of Bioethics

Cleveland Clinic

Cleveland, OH

Linda J. Young, Ph.D.

Professor of Statistics

Department of Statistics

Institute of Food and Agricultural Sciences

University of Florida

Gainesville, FL

* Not in attendance

** Attended via telephone

Attachment B

Federal Register Notice Announcing Meeting

Human Studies Review Board; Notice of Public Meeting   

[Federal Register: May 30, 2008 (Volume 73, Number 105)]

[Notices]

[Page 31117-31118]

From the Federal Register Online via GPO Access [wais.access.gpo.gov]

[DOCID:fr30my08-49]

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

ENVIRONMENTAL PROTECTION AGENCY 

[EPA–HQ–ORD–2008–0035; FRL–8573–9] 

Human Studies Review Board; Notice of Public Meeting 

AGENCY: Environmental Protection Agency (EPA). 

ACTION: Notice.

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

SUMMARY: The U.S. Environmental Protection Agency's (EPA or Agency)
Office of the Science Advisor (OSA) announces a public meeting of the
Human Studies Review Board (HSRB) to advise the Agency on EPA's
scientific and ethical review of human subjects research.

    Dates: The public meeting will be held from June 24-June 25, 2008
from 8:30 a.m. to approximately 5:30 p.m., Eastern Time (However, the
second day may not be needed).

    Location: Environmental Protection Agency, Conference Center—Lobby
Level, One Potomac Yard (South Bldg.), 2777 S. Crystal Drive, Arlington,
VA 22202.

    Meeting Access: Seating at the meeting will be on a first-come
basis. To request accommodation of a disability please contact the
person listed under FOR FURTHER INFORMATION CONTACT at least 10 business
days prior to the meeting, to allow EPA as much time as possible to
process your request.

    Procedures for Providing Public Input: Interested members of the
public may submit relevant written or oral comments for the HSRB to
consider during the advisory process. Additional information concerning
submission of relevant written or oral comments is provided in Unit I.D.
of this notice.

ADDRESSES: Submit your written comments, identified by Docket ID No.
EPA-HQ-ORD-2008-0355, by any of the following methods: 

    Internet: http://www.regulations.gov: Follow the on-line
instructions for submitting comments.

    E-mail: ORD.Docket@epa.gov.

    USPS Mail: Environmental Protection Agency, EPA Docket Center
(EPA/DC), ORD Docket, Mailcode: 28221T, 1200 Pennsylvania Ave., NW.,
Washington, DC 20460.

    Hand or Courier Delivery: The EPA/DC Public Reading Room is located
in the EPA Headquarters Library, Room Number 3334 in the EPA West
Building, located at 1301 Constitution Ave., NW., Washington DC. The
hours of operation are 8:30 a.m. to 4:30 p.m. Eastern Standard Time
(EST), Monday through Friday, excluding Federal holidays. Please call
(202) 566-1744 or e-mail the ORD Docket at ord.docket@epa.gov for
instructions. Updates to Public Reading Room access are available on the
Web site (http://www.epa.gov/epahome/dockets.htm).

    Instructions: Direct your comments to Docket ID No.
EPA-HQ-ORD-2008-0355. EPA's policy is that all comments received will be
included in the public docket without change and may be made available
online at http://www.regulations.gov, including any personal information
provided, unless the comment includes information claimed to be
Confidential Business Information (CBI) or other information whose
disclosure is restricted by statute. Do not submit information through
http://www.regulations.gov or e-mail that you consider to be CBI or
otherwise protected from disclosure. The http://www.regulations.gov Web
site is an ``anonymous access'' system, which means EPA will not know
your identity or contact information unless you provide it in the body
of your comment. If you send an e-mail comment directly to EPA, without
going through http://www.regulations.gov, your e-mail address will be
automatically captured and included as part of the comment that is
placed in the public docket and made available on the Internet. If you
submit an electronic comment, EPA recommends that you include your name
and other contact information in the body of your comment and with any
disk or CD-ROM you submit. If EPA cannot read your comment due to
technical difficulties and cannot contact you for clarification, EPA may
not be able to consider your comment. Electronic files should avoid the
use of special characters, any form of encryption, and be free of any
defects or viruses.

FOR FURTHER INFORMATION CONTACT: Any member of the public who wishes
further information should contact Lu-Ann Kleibacker, EPA, Office of the
Science Advisor, (8105R), Environmental Protection Agency, 1200
Pennsylvania Ave., NW., Washington, DC 20460; telephone number: (202)
564-7189; fax: (202) 564-2070; e-mail address:
kleibacker.lu-ann@epa.gov. General information concerning the EPA HSRB
can be found on the EPA Web site at http://www.epa.gov/osa/hsrb/.

SUPPLEMENTARY INFORMATION:

I. Public Meeting

A. Does This Action Apply to Me?

    This action is directed to the public in general. This action may,
however, be of particular interest to persons who conduct or assess
human studies, especially studies on substances regulated by EPA and to
persons who may sponsor or conduct research with human subjects with the
intention to submit it to EPA for consideration under the Federal
Insecticide, Fungicide, and Rodenticide Act (FIFRA) or section 408 under
the Federal Food, Drug, and Cosmetic Act (FFDCA). Since other entities
may also be interested, the Agency has not attempted to describe all the
specific entities that may be affected by this action. If you have any
questions regarding the applicability of this action to a particular
entity, consult the person listed under FOR FURTHER INFORMATION CONTACT.

B. How Can I Access Electronic Copies of This Document and Other Related
Information?

    You may access this Federal Register document electronically either
through http://www.regulations.gov or through the EPA Web site under the
Federal Register listings at http://www.epa.gov/fedrgstr/.

    Docket: All documents in the docket are listed in the
http://www.regulations.gov index under the docket number. Even though it
will be listed by title in the index, some information is not publicly
available, e.g., CBI or other information whose disclosure is restricted
by statute. Copyright material will be publicly available only in hard
copy. Publicly available docket materials are available either
electronically in http://www.regulations.gov or in hard copy at the ORD
Docket, EPA/DC, Public Reading Room. The EPA/DC Public Reading Room is
located in the EPA Headquarters Library, Room Number 3334 in the EPA
West Building, located at 1301 Constitution Ave., NW., Washington, DC.
The hours of operation are 8:30 a.m. to 4:30 p.m. EST, Monday through
Friday, excluding Federal holidays. Please call (202) 566-1744 or e-mail
the ORD Docket at ord.docket@epa.gov for instructions. Updates to Public
Reading Room access are available on the Web site
(http://www.epa.gov/epahome/dockets.htm). EPA's position paper(s),
charge/questions to the HSRB, and the meeting agenda are anticipated to
be available by late May 2008, if not earlier. In addition, the Agency
may provide additional background documents as the materials become
available. You may obtain electronic copies of these documents, and
certain other related documents that might be available electronically,
from the regulations.gov Web site and the HSRB Web site at
http://www.epa.gov/osa/hsrb/. For questions on document availability or
if you do not have access to the Internet, consult the person listed
under FOR FURTHER INFORMATION.

C. What Should I Consider as I Prepare My Comments for EPA?

    You may find the following suggestions helpful for preparing your
comments:

    a. Explain your views as clearly as possible.

    b. Describe any assumptions that you used.

    c. Provide copies of any technical information and/or data you used
that support your views.

    d. Provide specific examples to illustrate your concerns and suggest
alternatives.

    e. To ensure proper receipt by EPA, be sure to identify the docket
ID number assigned to this action in the subject line on the first page
of your response. You may also provide the name, date, and Federal

Register citation.

D. How May I Participate in This Meeting?

    You may participate in this meeting by following the instructions in
this section. To ensure proper receipt by EPA, it is imperative that you
identify docket ID number EPA-HQ-ORD-2008-0355 in the subject line on
the first page of your request.

    a. Oral comments. Requests to present oral comments will be accepted
up to June 17, 2008. To the extent that time permits, interested persons
who have not pre-registered may be permitted by the

Chair of the HSRB to present oral comments at the meeting. Each
individual or group wishing to make brief oral comments to the HSRB is
strongly advised to submit their request (preferably via e-mail) to the
person listed under FOR FURTHER INFORMATION CONTACT no later than noon,
Eastern Time, June 17, 2008 in order to be included on the meeting
agenda and to provide sufficient time for the HSRB Chair and HSRB
Designated Federal Officer (DFO) to review the agenda to provide an
appropriate public comment period. The request should identify the name
of the individual making the presentation, the organization (if any) the
individual will represent, and any requirements for audiovisual
equipment (e.g., overhead projector, LCD projector, chalkboard). Oral
comments before the HSRB are limited to five minutes per individual or
organization. Please note that this limit applies to the cumulative time
used by all individuals appearing either as part of, or on behalf of an
organization. While it is our intent to hear a full range of oral
comments on the science and ethics issues under discussion, it is not
our intent to permit organizations to expand these time limitations by
having multiple individuals sign up separately to speak on their behalf.
Each speaker should bring 25 copies of his or her comments and
presentation slides for distribution to the HSRB at the meeting. At the
discretion of the Board Chair and DFO, public commenters, if present
during the Board's discussion, may be asked to provide clarification of
their comments to assist the Board in their discussion.

    b. Written comments. Although you may submit written comments at any
time, for the HSRB to have the best opportunity to review and consider
your comments as it deliberates on its report, you should submit your
comments at least five business days prior to the beginning of the
meeting. If you submit comments after this date, those comments will be
provided to the Board members, but you should recognize that the Board
members may not have adequate time to consider those comments prior to
making a decision. Thus, if you plan to submit written comments, the
Agency strongly encourages you to submit such comments no later than
noon, Eastern Time, June 17, 2008. You should submit your comments using
the instructions in Unit I.C. of this notice. In addition, the Agency
also requests that person(s) submitting comments directly to the docket
also provide a copy of their comments to the person listed under FOR
FURTHER INFORMATION CONTACT. There is no limit on the length of written
comments for consideration by the HSRB.

 E. Background

1. Human Studies Review Board

    The HSRB is a Federal advisory committee operating in accordance
with the Federal Advisory Committee Act (FACA) 5 U.S.C. App.2 Sec.  9.
The HSRB provides advice, information, and recommendations to EPA on
issues related to scientific and ethical aspects of human subjects
research. The major objectives of the HSRB are to provide advice and
recommendations on: (a) Research proposals and protocols; (b) reports of
completed research with human subjects; and (c) how to strengthen EPA's
programs for protection of human subjects of research. The HSRB reports
to the EPA Administrator through EPA's Science Advisor.

2. Topics for Discussion

    For this meeting of the HSRB, the Board will present for HSRB review
scientific and ethical issues surrounding:

The Governing Document, the compilation of Standard Operating
Procedures, the scenario design document, and two associated protocols
from the Agricultural Handlers Exposure Task Force (AHETF), which
collectively describe research to monitor exposure of subjects who apply
an agricultural pesticide using airblast equipment in closed cabs.

A report from a completed laboratory study to evaluate the efficacy in
repelling mosquitoes of the genus Culex of two registered products
containing picaridin.

    The Board may also be reviewing its draft April 9-10, 2008 meeting
report for subsequent Board approval. In addition, the HSRB may also
discuss planning for future HSRB meetings.

3. Meeting Minutes and Reports

    Minutes of the meeting, summarizing the matters discussed and
recommendations, if any, made by the advisory committee regarding such
matters will be released within 90 calendar days of the meeting. Such

minutes will be available at http://www.epa.gov/osa/hsrb/ and
http://www.regulations.gov. In addition, information concerning a Board
meeting report, if applicable, can be found at http://www.epa.gov/osa/

hsrb/ or from the person listed under FOR FURTHER INFORMATION CONTACT.

Dated: May 23, 2008.

George Gray,

EPA Science Advisor.

[FR Doc. E8-12144 Filed 5-29-08; 8:45 am]

BILLING CODE 6560-50-P

Attachment C

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

HUMAN STUDIES REVIEW BOARD (HSRB)

JUNE 24-25, 2008 *

PUBLIC MEETING

Tuesday, June 24, 2008

Environmental Protection Agency

Conference Center – Lobby Level

One Potomac Yard (South Bldg.)

2777 S. Crystal Drive

Arlington, VA 22202

HSRB WEB SITE   http://www.epa.gov/osa/hsrb/

Docket Telephone: (202) 566-1752

Docket Number: EPA-HQ-ORD-2008-0355

8:30 AM	Convene Meeting and Identification of Board Members – Celia
Fisher, Ph.D. (HSRB Chair)

8:40 AM	Welcome – Elizabeth Lee Hofmann, Ph.D. (Deputy Director,
Office of the Science Advisor [OSA])  

8:50 AM	Opening Remarks – Debbie Edwards, Ph.D. (Director, Office of
Pesticide Programs [OPP], EPA)

9:00 AM	Meeting Administrative Procedures – Paul Lewis, Ph.D.
(Designated Federal Officer [DFO], HSRB, OSA, EPA) 

9:10 AM	EPA Follow-up on Pesticide Specific HSRB Recommendations – Mr.
William Jordan (OPP, EPA)

AHETF Pesticide Handler Protocols: Closed-Cab Airblast Scenario

9:30 AM	Overview – Mr. William Jordan (OPP, EPA)

9:45 AM	Science and Ethics Assessment of AHETF Protocol – Mr. Jeffrey
Evans (OPP, EPA), Mr. Matthew Crowley (OPP, EPA) and Mr. John Carley
(OPP, EPA)    

10:30 AM	Break    

10:45 AM	Science and Ethics Assessment of AHETF Protocol (continued) –
Mr. Jeffrey Evans (OPP, EPA), Mr. Matthew Crowley (OPP, EPA) and Mr.
John Carley (OPP, EPA)    

11:30 AM	Public Comments    

12:00 PM	Lunch   

1:00 PM	Board Discussion    

If AHETF’s proposed closed-cab airblast application scenario design,
field study protocols AHE55 and AHE56, and supporting documentation are
revised as suggested in EPA’s reviews:

Does the research appear likely to generate scientifically reliable
data, useful for assessing the exposure of handlers who apply liquid
pesticides using airblast equipment drawn by vehicles with closed cabs? 

Does the research appear to meet the applicable requirements of 40 CFR
part 26, subparts K and L?  

3:30 PM	Break  

Completed ICR Mosquito Repellent Efficacy Study A117

3:45 AM	EPA Science and Ethics Assessment of Completed ICR Mosquito
Repellent Efficacy Study A117 – Mr. Mark Suarez (OPP, EPA) and Mr.
John Carley (OPP, EPA) 

4:45 PM	Public Comments 

5:15 PM	Adjournment – Celia Fisher, Ph.D. (HSRB Chair) and Paul Lewis,
Ph.D. (HSRB DFO)

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

HUMAN STUDIES REVIEW BOARD (HSRB)

JUNE 24-25, 2008 *

PUBLIC MEETING

Wednesday, June 25, 2008

Environmental Protection Agency

Conference Center – Lobby Level

One Potomac Yard (South Bldg.)

2777 S. Crystal Drive

Arlington, VA 22202

8:30 AM	Convene Meeting – Celia Fisher, Ph.D. (HSRB Chair)

8:35 AM	Follow-up from Previous Day’s Discussion – Mr. William
Jordan (OPP, EPA)  

Completed ICR Mosquito Repellent Efficacy Study A117

8:45 AM	Board Discussion        

Is this study sufficiently sound, from a scientific perspective, to be
used, in conjunction with other information, to assess the repellent
efficacy of the formulations tested against mosquitoes of the genus
Culex?     

Does available information support a determination that this study was
conducted in substantial compliance with subparts K and L of EPA
regulations at 40 CFR Part 26?  

Review of April 9-10, 2008 HSRB Meeting Report

9:45 AM	Review Process – Celia Fisher, Ph.D. (HSRB Chair)

9:55 AM	Public Comments

10:15 AM	Break  

10:30 AM	Board Discussion and Decision on Report – Celia Fisher, Ph.D.
(HSRB Chair)   

12:30 PM	Concluding Remarks – Mr. William Jordan (OPP, EPA)  

12:35 PM	Adjournment – Celia Fisher, Ph.D. (HSRB Chair) and Paul
Lewis, Ph.D. (HSRB DFO)

* Please be advised that agenda times are approximate and subject to
change. For further information, please contact the Designated Federal
Officer for this meeting, Paul Lewis, via telephone: (202) 564-8381 or
email: lewis.paul@epa.gov.

  PAGE  1  of   NUMPAGES  69 

