UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON D.C., 20460
OFFICE OF CHEMICAL
SAFETY AND POLLUTION
PREVENTION
March 30, 2018
MEMORANDUM
SUBJECT: Science and Ethics Review of a Protocol for Laboratory Evaluation of Skin-
Applied Tick Repellent Product Containing OLE
FROM: Clara Fuentes, Ph.D., Entomologist
Biopesticides and Pollution Prevention Division
Office of Pesticide Programs
Eric Bohnenblust, Ph.D., Entomologist
Biopesticides and Pollution Prevention Division
Office of Pesticide Programs
Michelle Arling, Human Research Ethics Review Officer
Office of the Director
Office of Pesticide Programs
TO: Linda Hollis, Chief, Biochemical Pesticides Branch
Biopesticide and Pollution Prevention Division
Office of Pesticide Programs
REF: Logan, James, Study Director. (2017) Protocol for "A single group trial to
determine the complete protection time of an insect repellent formulation
containing 30% Citriodiol® (Oil of Lemon Eucalyptus) against three species of
ticks." Unpublished document prepared by Citrefine International Ltd., Moorfield
Road, Yeadon, Leeds, LS19 7BN, United Kingdom. 21st July 2017. 37 p. MRID
504422-01.
We have reviewed the referenced protocol for laboratory testing for a skin-applied
repellent product containing Oil of Lemon Eucalyptus (OLE) against three species of ticks in a
laboratory setting from both scientific and ethics perspectives. This protocol was submitted by
arctec (Arthropod Control Product Test Centre), part of the London School of Hygiene and
Tropical Medicine (LSHTM). The study is sponsored by Citrefine International, Ltd. This review
assesses the scientific aspects of the proposed research for a product performance study to evaluate
the efficacy of skin applied insect repellent products in terms of the recommendations of the EPA
OCSPP 810.3700 Guideline, Insect Repellents to be Applied to Human Skin, and the EPA Human
Studies Review Board (HSRB), concerning scientific merit of the proposed study. Ethical aspects of
w
PROlt&
T>
1.
ID
-------
the proposed research are assessed in terms of the standards defined by 40 CFR 26 subparts K
and L.
A. Completeness of Protocol Submission
The submitted protocol was reviewed for completeness against the required elements listed
in 40 CFR §26.1125. EPA's checklist is appended to this review. All elements of required
documentation are provided in the submitted protocol package and supplementary documentation
of review by Western Institutional Review Board (WIRB).
B. Summary Assessment of Ethical Aspects of the Proposed Research
Here is a summary of our observations about the ethical aspects of the proposed protocol
as amended to address EPA's comments and recommendations. EPA and arctec met in March to
discuss EPA's review and recommendations; arctec reviewed and agreed to make all suggested
changes, except deleting the consumer dose phase and using EPA's recommended standard dose
during the repellency testing portion of the study. Attachment 1 provides supporting details and a
point-by-point evaluation of this protocol.
1. Societal Value of Proposed Research: This study is designed to determine the
efficacy and protection time of a topically topically-applied skin repellent against ticks.
Efficacy will be expressed as Complete Protection Time (CPT), which is defined as the
time between application of the repellent product and the time at which 1 tick crosses 3
cm of treated skin, followed within 30 minutes by a second confirmatory crossing. The
study will also include a dosimetry phase to establish an expected "consumer dose" that
will be used as the standard dose during the repellent efficacy trial. The research has
societal value because people are at risk of contracting tick-borne diseases, and such
risks can be mitigated by the use of insect repellent products. There are no data
showing the necessary efficacy of OLE in human studies and no recommended dose
for the proposed delivery mechanism for this product. The rationale for this testing is
to collect data to estimate the CPT for this tick repellent product containing OLE. As
intended, the data resulting from this proposed study will be used to support
registration of topically-applied repellents containing OLE.
2. Subject Selection: With EPA's comments and recommendations incorporated,
subjects will be recruited through emails, posters, and social media targeted to the
London area, namely from the area in and around the London School of Hygiene and
Tropical Medicine. The advertisements will be posted on bulletin boards and
electronically in order to ensure that there is equity of access to participate in the study.
The advertisement will provide a brief description of the study and an email address
and phone number that interested persons can use to contact the arctec Trial
Coordination Centre to get more information about the study and how to participate.
Study staff will contact persons interested in participating by phone or email to provide
additional information about the study and to determine whether they meet basic
inclusion and exclusion criteria. Potentially qualified persons will be added to a
recruitment list in the order of their enrollment. Once a pool of at least 75 potentially
Page 2 of 70
-------
qualified persons is assembled, the list will be randomized. The potentially qualified
subjects will be contacted in order of the randomized list and invited to the arctec
facility for a one-on-one consent meeting with a member of the study team.
The study will be conducted in two phases: a dosimetry phase, to determine a typical
consumer dose, and the tick repellent efficacy testing phase. A total of 25 subjects will
be necessary to complete the dosimetry phase. For the repellent efficacy testing phase,
EPA recommends a test subject sample size of 25 for each of the three tick species that
will be tested to generate statistically-sound data (see Attachment 3). As discussed
above, the subjects will be contacted according to the order of the randomized list of
qualified persons. At the end of the dosimetry phase, staff will confirm with enrolled
and consented subjects whether they wish to participate in testing the repellent against
one or more tick species. As necessary, additional subjects from the generated
randomized list will be called on to enroll in testing against one or more tick species as
necessary to ensure that a total of 25 subjects complete a repellency efficacy trial
against each of the three species of ticks discussed in this protocol.
If the original pool of 75 subjects is not sufficiently large to complete the dosimetry
phase and repellent efficacy trials against all 3 species of ticks, recruitment will be re-
opened as described above until another pool of at least 25 potentially qualified
subjects has been assembled. Then the same randomization process will be followed to
add them to the end of the previously compiled randomized list. Then the potentially
qualified subjects will be invited for a consent meeting and, if qualified, enrolled into
the study in sequential order.
The inclusion and exclusion criteria, with EPA's recommendations addressed, are
appropriate and complete.
3. Risks to Subjects: The protocol discusses potential hazards associated with these
tests including adverse reactions to the test substance, adverse reaction to tick bites,
potential transmission of tick-borne disease, potential stress from finding out the
results of a pregnancy test, and unintentional release of confidential information. The
protocol notes that risks will be minimized as follows. To minimize risks of adverse
reactions to the test substance, subjects will be enrolled only if they do not have a
known sensitivity or allergy to the test substance or any skin-applied insect repellents.
In addition, subjects with localized skin disorders on the forearms that could be
exacerbated by exposure to the test substance will be excluded.
To mitigate risks from exposure to ticks, ticks will be placed one at a time on subjects'
arms and monitored closely to minimize the chance for the tick to move further onto
the host for a suitable location to attach (e.g., armpit). Ticks will be removed if they
begin to bite and attach to the subject. Should a tick exhibit behavior indicating it is
attaching to the subject, it will be removed immediately by the researcher with fine
forceps. To eliminate the risk of transmission of tick-borne disease, ticks will be
sourced from pathogen-free colonies at Oklahoma State University Tick Rearing
Facility. Ticks will only be used once with a single subject, and ticks will be destroyed
by immersion in alcohol after they are placed on a subject's arm.
Page 3 of 70
-------
EPA's suggested edits to the protocol incorporate additional protections to keep the
subjects' identities and results of pregnancy testing private. Practical steps to minimize
subject risks have been described in the protocol, and the remaining risks have a low
probability of occurrence.
With EPA's recommendations addressed, the risks to subjects have been identified
and appropriate steps to minimize risks are included.
4. Benefits: This research offers no benefits to subjects. Depending on the results of the
research, it may provide indirect benefits to subjects and society by potentially leading
to data that could be used by EPA to register insect repellent products containing OLE.
These repellent products could lead to fewer tick bites and reduced incidents of tick-
borne illnesses.
5. Risk/Benefit Balance: The protocol describes measures to further reduce risk to subjects
while maintaining the robustness of the scientific design. Due to the risk mitigation
measures put in place, the residual risk to subjects is low and reasonable in light of the
potential benefits of the data to society.
Subjects will be exposed to the test substance in two phases - the dosimetry phase,
and the repellent efficacy phase. Although EPA recommends using a standard dose for
repellent efficacy trials to avoid additional exposure of subjects during a consumer
dose phase, OPPTS Guidelines 810.3700 provides guidance on conducting a "dose
determination" phase to establish a typical consumer dose for use in a subsequent
repellent efficacy trial (pp. 23-25). The Study Sponsor believes that the typical
consumer dose for their product is higher than EPA's recommended standard dose
(0.5 g/600 cm2). Testing the product at a lower rate than consumers would typically
use could result in an underestimation of the product's efficacy, which could result in
consumers over-applying the product. The product being tested has already approved
by EPA and has been on the market for more than a decade, the product is only being
applied to a fraction of the skin that it would be in normal use, and the product would
be removed immediately after each application so exposure is further limited. The
minor incremental risks of additional exposure to subjects participating in the
dosimetry phase are reasonable when considered with the potential benefits of
ensuring the product labeling provides an estimate of protection time based on
consumer practices and of avoiding over-application of the product by future
consumers. When set against the far greater risk of consumer over application which
would result from an underestimate of the product's efficacy in preventing tick bites,
it suggests that the dose rate study is fully justified.
6. Independent Ethics Review: Western Institutional Review Board (IRB) has reviewed
and approved the protocol, informed consent form, and recruitment materials. Western
IRB is registered with the Office of Human Research Protections at the Department of
Health and Human Services (IRB00000533). Western IRB has been accredited by the
Association for the Accreditation of Human Research Protection Programs (AAHRPP)
since 2003. Satisfactory documentation of the Western IRB procedures and
Page 4 of 70
-------
membership is on file with the Agency and has been provided to the HSRB members.
Documentation regarding IRB approval of the protocol has been provided to the HSRB
members with the background materials for this protocol.
7. Informed Consent: With the agency's comments addressed, the protocol contains a
complete and satisfactory description of the process by which potential subjects will be
recruited, informed and trained in preparation for the test day, and the process for
seeking subjects' consent to participate. A copy of the IRB-approved consent document
is included in the background materials; with EPA's recommendations and comments
incorporated, it meets the requirements of 40 CFR §§26.1116 and 26.1117
8. Respect for Subjects: The subjects' identities will be protected as follows: each
subject will be assigned a code number/identifier. The study records will be maintained
in locked cabinets, and electronic files kept on a password-protected computer server
or encrypted electronic storage devices. The protocol will be revised to explain the
provisions made for discrete handling of the pregnancy testing that is required of
female subjects on each day of testing. Any photographs or videos of the subjects will not
include their faces, tattoos, or other identifying features. Candidates and subjects will be
informed that they are free to decline to participate or to withdraw at any time for any
reason. Subjects will be compensated as described in the protocol. Breaks for subjects
between exposures and provision of snacks and drinks for interested subjects will be
incorporated into the study design.
C. Compliance with Applicable Ethical Standards
This is a protocol 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. Thus the primary ethical standards applicable to this proposal are 40 CFR 26,
Subparts K and L. In addition, the requirements of FIFRA §12(a)(2)(P) for fully informed, fully
voluntary consent of subjects apply. A point-by-point evaluation of how this protocol addresses
the requirements of 40 CFR 26 Subparts K and L and the criteria recommended by the HSRB is
appended as Attachment 1.
Although this research will take place in a foreign country, the Study Sponsor has agreed
to conduct the research in compliance with the requirements of 40 CFR 26 Subpart K and L. In
addition, the research will comply with the World Medical Association's Declaration of Helsinki
(2003), which covers the principles of respect for individuals, beneficence, and justice. These are
also the foundations of United States' regulations for the protection of human subjects.
EPA's Ethics Comments
The London School of Hygiene and Tropical Medicine/arctec were notified that, before the
research is conducted, the protocol and supporting documents must be revised to address EPA's
comments and recommendations resulting from the review by the Human Studies Review Board
(HSRB). They have already agreed to address EPA's comments. To facilitate the HSRB's review of
the latest protocol, which incorporates the EPA's comments, the EPA is providing a separate file for the
HSRB titled "OLE Revised Protocol with EPA Comments Incorporated." After the HSRB
Page 5 of 70
-------
completes its review of the protocol and relays its recommendations to the EPA, the EPA and
London School of Hygiene and Tropical Medicine/arctec will reach agreement on
implementation of the HSRB's recommendations; the revised protocol and supporting documents
should be resubmitted for review and approval to Western IRB prior to initiating the research.
The EPA's ethics comments are provided below and presented in the order they are found
in the protocol. Minor editorial comments and corrections of typographical errors have not been
included here. In addition, EPA has provided ethics-related comments on the informed consent
materials; these are provided to the HSRB as a separate file.
General Comments & Study Synopsis
1. Add the date of IRB approval to the protocol.
2. Revise the subject size throughout the protocol to reflect the statistically-supported
sample size of 25 subjects for each of the tick repellent efficacy tests. In addition,
indicate that 25 subjects will be enrolled for the dosimetry phase. Ensure that the
protocol and consent form clearly state that subjects may enroll in one or more test
days. Ensure that at least 10 subjects of each sex are recruited for each test, with a
goal of having at least 12 subjects of each sex.
3. Revise the protocol to clearly state that at least 48 hours will elapse between a
subject's participation in a 2nd test day; based on discussions between the researchers
and EPA, a 1 week waiting period is unnecessary. Provide rationale for follow up at
48 hours and 1 week.
4. Separate "eligibility criteria" into "inclusion criteria" and "exclusion criteria".
5. Revise exclusion criteria to include the following:
•
• Known phobia of ticks or tick bites
• Employees and spouses of employees of the Study Sponsor (Citrefine, Inc.),
arctec, and those directly line managed by Professor James Logan (study
director)
• Students of the study director, principal investigator, or any other faculty
members involved in this study
Section 4 - Trial Design
6. In Section 4.2, Risks and Benefits Assessment:
• Separate out each risk and discuss the corresponding benefits.
• Regarding potential for eye injury, please have eye wash available for
subjects in the event they get the product in their eyes.
• Provide specific information on how ticks will be destroyed.
• Clarify whether tick paralysis is an actual risk to subjects if the ticks will not
be allowed to attach/bite.
• Again, revise period between a subject's participation to at least 48 hours.
• Add risk of unanticipated loss of confidential information and psychological
risks associated with pregnancy testing, and corresponding risk mitigation
measures.
Section 5 - Participant Entry
Page 6 of 70
-------
7. In Section 5.1, Recruitment:
• Revise the estimated maximum number of subjects needed for the study to
reflect the revised number of subjects for the dosimetry and repellent efficacy
phases.
• Expand recruitment to the London area - limiting recruitment to the London
School of Hygiene and Tropical Medicine is too narrow and could result in a
skewed pool of subjects.
• Provide to EPA copies of the materials that will be used during recruitment,
e.g., Product Information Sheet, sample email and telephone script that will
be used to follow up with those interested in participating.
• Include a randomization process to avoid biasing the pool towards those who
respond immediately. (Note: This could also be included in Section 6.2.) For
example:
"Recruitment will initially be open for a 2-week period. The goal is to have at
least 75 potentially qualified subjects. If this target is not reached after 2
weeks, recruitment will continue for an additional week at a time, until the
target is reached. The total number of qualified subjects will each be
assigned a unique and consecutive number, starting at OLE-Ol based on the
order of their enrollment. The numbers will then be randomized using the
randomize function in Microsoft Excel. The first 25 subjects in the generated
randomized list will be consented for the dosimetry phase and will be offered
the opportunity to consent for one or more repellent efficacy trials. If needed,
subjects from the generated randomized list will be called on in sequential
order to replace any subject who withdraws from the dosimetry phase.
"At the end of the dosimetry phase, staff will confirm with enrolled and
consented subjects whether they wish to participate in testing the repellent
against one or more tick species. As necessary, additional subjects from the
generated randomized list will be called on to enroll in testing against one or
more tick species as necessary to ensure that a total of 25 subjects complete a
repellency efficacy trial against each of the three species of ticks discussed in
this protocol.
"If the original pool of 75 subjects is not sufficiently large to complete the
dosimetry phase and repellent efficacy trials against all 3 species of ticks,
recruitment will be re-opened as described above until another pool of at
least 10 potentially qualified subjects has been assembled. Then the same
randomization process will be followed to add them to the end of the
previously compiled randomized list and if they are qualified and interested,
they will be enrolled into the study in sequential order. "
8. For Section 5.2, Eligibility Criteria:
• Separate the list into inclusion and exclusion criteria.
• Add to the inclusion criteria "Non-smoker or willing to refrain from smoking
for 24 hours prior to and during each test".
• Revise the exclusion criteria listed below as follows:
- Known or suspected history of tick bite paralysis or tick bite allergies
- Known phobia of ticks or tick bites
Page 7 of 70
-------
- Known allergy or sensitivity to Oil of Lemon Eucalyptus, any other
repellent ingredients, or skin-applied repellent products
- Employees and spouses of employees of the Study Sponsor (Citrefine,
Inc.), arctec, and those directly line managed by Professor James Logan
(study director)
Students of the study director, principal investigator, or any other faculty
members involved in this study
9. Clarify that non-scented soap will be provided to subjects at the consent meeting.
10. For Section 5.3, Consent Process:
• Provide a thorough description of what will be covered during the consent
meeting, including a brief outline of the study including its purpose, the
subjects' potential role in the study, the potential length of the study on any
given test day, the identity and function of the repellent to which subjects will
be exposed, the potential hazards associated with the study and steps being
taken to mitigate each hazard as addressed in the protocol, and the
inclusion/exclusion criteria. The procedures involved with the dosimetry test
should be demonstrated step-bv-step to all subjects who participate in the
training prior to the dosimetry phase. The study team member should explain
and demonstrate how the test substance will be applied and give a step-by-
step description of how a 15-minute exposure interval during the tick
repellent efficacy portion of the trial will be conducted.
• Pregnancy testing on each day of participation must be explained to female
subjects. This discussion should cover how and when testing will occur and
how the female subject's privacy will be respected.
• Revise consent process as necessary to reflect recruitment and enrollment in
both dosimetry and repellent efficacy phases.
• Ensure that subjects are informed that they can speak with the Study Director
in private during or after the consent meeting if they have any questions.
• Include questions that will be asked by the person conducting the consent
meeting to ensure that subjects demonstrate comprehension of the materials
presented prior to giving informed consent to enroll in the study.
• Revise this section to note that pregnancy testing must be conducted on each
day of the study.
11. In Section 5.4, Withdrawal Criteria, add the following language immediately
following the first sentence: "Participants will be told during the consent process and
again at the initiation of each test day in which they participate that they may
withdraw at any time without giving a reason and without forfeiting benefits to which
they are entitled."
Section 6 - Intervention
12. Separate the discussion of the interventions into a description of the test substance,
randomization, dosimetry phase, and repellent efficacy phase.
Section 7 - Participant Timeline
13. Revise this section to align with all other edits to the protocol (timing, recruitment,
enrollment, number of subjects, follow up/waiting periods).
Page 8 of 70
-------
Section 8 - Test Methodology
14. Revise instructions to participants for dosimetry phase. Subjects should be provided
the label and instructed to read the label and apply according to the instructions.
Researchers should not influence the amount of product applied during the dosimetry
phase.
15. Explain when and how the subjects for the dosimetry phase will be provided with
safety glasses/goggles to protect their eyes during application.
16. Prior to initiating the dosimetry phase, both forearms of each participant should be
washed with unscented soap and water and dried with paper towels.
17. State explicitly that on each day of testing, females will complete a pregnancy test
prior to being exposed to the test substance.
18. Prior to both phases, a study team member or medical professional should inspect the
subject's forearms for disqualifying skin conditions. Include this step in the protocol
and consent form.
19. For Section 8.2, Assessment of complete protection time:
• Provide information about how subjects' comfort will be accommodated
during the testing. For example, provide information about how/where
subjects will be seated, and that adequate time will be allowed between trials
for subjects to stand and stretch, to use the restroom, to eat or drink, etc.
Section 10 - Safety Reporting and Data Monitoring
20. Explain how the medical monitor will participate in the study. Will he be on-site or
on-call during monitoring events? How will he respond in the event of an adverse
effect? Ensure that the medical monitor is familiar with the protocol prior to study
initiation.
21. Provide on-call first aider a copy of final approved protocol, brief them on study
process and test substances, and if applicable, contact on-call first aider at initiation of
each test day to confirm that testing has begun for that day.
22. Provide to subjects during the consent meeting instructions on how to report adverse
events as described in this section.
23. For reporting adverse events, ensure that the protocol includes language specific to
the reporting requirements for Western IRB as well.
Section 11 - Ethics & Dissemination
24. In section 11.1, include WIRB requirements for reporting changes to the protocol
made to eliminate apparent immediate hazards to human participants, e.g., must be
reported in writing within X days of the change.
25. In section 11.2, include the following language: "All deviations (including minor
corrections) will be documented by the Study Director, reported to the IRB as
required, and included in the final study report provided to EPA."
26. Revise section 11.3 as follows: "Consent to enter the study must be sought from each
volunteer only after a full explanation has been given, an information leaflet offered
and time allowed for consideration. Signed volunteer consent should must be
obtained. The right of the participant to refuse to participate without giving reasons
must be respected. Participants must be informed of their right to withdraw from
participation at any point in the study without forfeiting benefits to which they are
entitled."
Page 9 of 70
-------
27. Compensation for participants should be revised. Participants should be compensated
for participating in an in-person screening appointment/consent meeting. Volunteers
in the tick trial, who could spend over 10 hours at the test facility, should be
compensated adequately for their inconvenience and loss of employment income.
EPA recommends compensating subjects for the consent meeting at a flat rate (e.g.,
£20 each for taking part in the screening and consent meeting, which could take up to
an hour). EPA also recommends compensating subjects on an hourly basis, at close to
the minimum wage (note: in the UK, minimum wage varies by age), £7.50/hour for
the lab-based repellent testing, rounded up to the next hour.
D. Summary Assessment of Scientific Aspects of the Proposed Research
1. Objectives
The main objectives of the proposed study are to determine the complete protection time
(CPT) of the skin-applied repellent products, EPA Reg. No. 84878-2 and 305-62,
containing 30% Citriodiol® (Oil of Lemon Eucalyptus (OLE); CAS No. 1245629-80-4;
PC Code: 040522) as the active ingredient, against ticks, and to determine the duration
of 90% repellency. The products are proposed to be tested on 3 tick species, Ixodes
scapularis, Amblyomma americanum, and Rhipicephalus sanguineus at the typical
consumer dose. The secondary objective is to estimate the average dose applied by
consumers (a.k.a. typical consumer dose) measured as ml/cm2. (Study Synopsis (pg. 3)
Section 1. a. Objectives (pg. 7)).
2. Endpoints
Efficacy endpoints are defined on pg. 4, Study Synopsis, as "Time to complete
protection time measured as a single tick crossing 3 cm into treated skin, and time to 90
% protective efficacy measured as more than 10% of ticks crossing treated skin.,,
Efficacy endpoints are defined on pg. 9, Section 4.1 Trial Endpoints, as "The primary
endpoint measure is the complete protection time (CPT), which is defined as the time
between application of the repellent product and the time at which 1 tick crosses 3cm of
treated skin, followed within 30 minutes by a second confirmatory crossing. The median
CPT will be calculated for each species using Kaplan-Meier survival curves. The first
secondary endpoint measure is the averase dose rate applied by consumers when using
EPA Reg. No. 84878-2 (alt) formulated insect repellent according to label instructions
for purposes of repelling ticks. Other secondary endpoint measures will be the time to
90% protective efficacy of EPA Reg. No. 84878-2 (alt) formulation against Ixodes
scapularis. Amblyomma americanum and Rhipicephalus sanguineus."
The endpoint for estimation of CPT is also defined on pg. 18, Section 9.2. Statistical
methods as the time to failure, which is the time when a first tick crossing into 3 cm of
treated skin is followed by a second confirmatory crossing by a different tick 30 minutes
Page 10 of 70
-------
apart from the first crossing. "The endpoint for CPT will be time to treatment failure for
each participant. Treatment failure is the time at which the product no longer provides
complete protection, which is determined as the time at which 1 tick crosses 3 cm of
treated skin, followed within 30 minutes by a second confirmatory crossing. "
The endpoint for dosimetry is defined in in Study Synopsis, pg. 4, and on pg. 9, Section
4.1 Trial Endpoints as "The typical consumer dose of one insect repellent product
measured as ml/cm2. " And It is also defined on pg. 18, Section 9.1. Statistical methods
as "The endpoint for the consumer dose assessment is the mean dose from all 21
volunteers in mg/cm2. This will be converted to ml/cm2 taking into account the specific
gravity of the test material. "
3. Sample size
The study protocol proposes to test the product on 30 participants (50:50 sex ratio),
which corresponds to a sample size of 10 subjects per tick spp. A sample size of 21
participants is proposed for determination of consumer dose. " Twenty-one participants
will be requiredfor the consumer dose evaluation whereas repellent product testing
requires up to 30 participants (preferably with a 50:50 ratio of males to females).
Participants will test the product at a single dose against at least one of the three [tick]
species." (pg. 8, Section 4. Trial Design).
" Twenty-one participants will be requiredfor the consumer dose study and ten
participants (at least three of each gender) will be required per tick species. All
participants will be 18 to 65 years of age. Participants can take part in multiple
experiments if they wish but can only do each test once. Therefore, a minimum of 21,
and a maximum of 51 volunteers will be required to complete the study. " "Volunteers
will be given the option to consent to the determination of the consumer dose and/or to
the determination of the CPT of one or more insect tick species. " (pg. 11, Section 5.
Participant Entry, and Subsection 5.1. Recruitment Number of Participants).
Justification for a sample size of 21 participants for determination of consumer dose is
provided on pg. 29, Section 14.2 Sample Size Calculations of the study protocol. The
level of precision is 0.1 |il/cm2 with a 95% confidence interval, for example 1.5-1.6
|il/cm2. The 1.6 |il/cm2 value is the standard deviation of doses applied to dosimeters
derived from previous studies (references not provided). The formula used to calculate
sample size at the given level of precision was taken from the publication of Kirkwood
and Stern (2003). According to the sample size of 61 subjects was corrected for
clustering as a single participant would contribute up to 9 data points (Hayes and
Bennet, 1999), resulting in a sample size of 21 participants (See Table Al, Sample size
calculations to determine the consumer dose with a specified level of precision on pg. 29
of the Study protocol, and Table 2 at the end of this review).
4. Study duration
Page 11 of 70
-------
The study is anticipated to last 12 weeks to have sufficient time to conduct screening,
dosimetry testing, and repellent testing against all three tick species. Prior to efficacy
testing, participants will undergo screening and dosimetry testing, which are estimated
to last 1 hour, followed by a follow up period of 48 hours. Efficacy testing is scheduled
for up to 10 hours from the time of product application, with follow ups of 48 hours and
1 week between test days (for adverse effects). (Pg. 13 Section 7. Participant timeline).
5. Determination of consumer dose and determination of CPT
The study is split into two parts; the first part is for determination of consumer dose, and
the second part is for determination of CPT. Efficacy testing is designed to assess the
longevity of consumer dose against three tick species. Consumer dose determination
(Dosimetry) will employ 21 volunteers; each volunteer will perform 3 product
applications to calculate the average dose per participant. The consumer dose, measured
in mg of applied product /cm2 skin surface area, is based on the grand mean of
triplicates application across 21 subjects. The consumer dose, measured in mg/cm2, will
be converted to ml/cm2 using the specific gravity of the formulation. Dosimetry phase is
described on pg. 15, Section 8. Test Methodology. Subsection 8.1. Consumer dose
determination. "Participants will be asked to carry out a practice application to one
forearm in order to allow them to get a feelfor the spray bottle andformulation. The
repellent will then be washed off with soap and water and patted dry with paper towels.
Three pre-weighed gauze bracelets 3cm wide will be spaced evenly along the
participant's forearm, and the arm circumference at these points recorded. Participants
will be asked to apply the repellent to their forearm only as if they were about to enter a
tick-endemic area. Two more applications will then be performed by each participant
with the arm washed between each repetition. The weight gain of the bracelets will be
used to determine the application rate applied. The "typical dose " in mg/cm2 [converted
to ml/cm2 skin, using specific gravity of the formulation] will be calculated as the
[overall or grand mean from the] mean of multiple applications [3 applications per
subject] by each of the participants and [converted to volume by] the specific gravity of
the test material."
5.1 Estimation of Skin Surface Area
The skin surface area of both forearms (left and right) will be calculated for every
subject forearms as the product of the sum of Wrist circumference + Elbow
circumference multiplied by V2 of the forearm Length. The procedure is described on pg.
15, Section 8.2. Assessment of complete protection time. "Participants' forearms (left
and right) will be measured, and the surface area calculated using the formula:
Forearm area = (Wrist circumference + Elbow circumference) x V2 Length forearm.
Page 12 of 70
-------
The dose required [for testing efficacy] will then be calculated based on the outcome of
the consumer dose test
6. Randomization
Treatment will be applied to either left or right forearm depending on a randomization
schedule. The schedule will randomly assign treatment to either left or right forearm for
each of the 3 tick species tested for each participant.
(pg. 13. Section 6.2. Randomization).
7. Dose Application
Consumer dose for testing efficacy will be applied by the same research assistant, using
a micropipette and applied with a single nitrile gloved finger to the whole right or left
forearm from the boundary line up to the elbow, as described on pg. 13, Section 6.
Interventions, "The repellent will be carefully rubbed evenly onto the skin using a single
finger by the same researcher wearing nitrile gloves. Participants will be asked to avoid
rubbing or washing the arm for the first 20 minutes after application. If accidental or
deliberate removal does occur, the test will be terminated for that day and repeated on
another."
8. Study Design
Efficacy testing for estimation of CPT will be conducted under laboratory conditions of
20-25 °C and over 35 % relative humidity. The treatment will be randomly applied to
either the left or right forearm of each participant for each test. Each subject will act as
their own control. There is no blinding employed since the outcome measures are based
on tick behavior, and no positive controls are proposed. Both forearms of each
participant, control and treated forearm, will be arranged in identical fashion, washed
with unscented soap and water, dried with paper towels, and marked with 3 lines, 3 cm
apart form each other, starting at the wrist toward the elbow. One line is drawn around
the wrist, the "release" line, another is drawn 3 cm above the release line toward the
elbow, that is the "boundary" line, marking the edge of the treated skin area, and a third
line, drawn 3 cm above the boundary line toward the elbow, that is the crossing line
(Figure 2 on pg. 16, Section 8.2. Assessment of complete protection time). These lines
are used to define the criteria for repellency. Ticks crawling beyond the boundary line
within 3 minutes and remaining on treated skin for 1 minute, are considered not
repelled. Ticks that do not cross the boundary line, or that do but turn back or fall off
are classified as repelled. Only ticks that are actively questing will be selected for
repellency testing. Ticks will be screened for testing by placing individual ticks,
oriented toward the elbow, at the release line on the untreated forearm. The untreated
forearm will be resting at a 30° angle on a flat surface. Ticks that crawl upward across
the boundary line toward the elbow will be selected for efficacy testing, and those that
fail will be discarded by immersing them in isopropyl alcohol. The same procedure will
Page 13 of 70
-------
be repeated on the treated arm with actively questing ticks. The movement of the tick on
the treated forearm will be timed for 3 minutes from the moment the tick is released on
the release line (pg. 16, Section 8.2. Assessment of complete protection time).
9. Statistical Analysis and Sample Size Calculation
The median CPT within 95% confidence intervals, will be estimated from the CPT for
each participant per tick species, using Kaplan-Meier survival analysis (pg. 18. Section
9. Statistical Analysis). Sample size calculations on pg. 18, under Section 9.1 Sample
Size Calculations, have been revised and amended according to EPA's power simulation
for determination of sample size in Attachment 3). The following is a brief summary of
EPA's comments on sample size determination:
• EPA obtained the raw data from the authors of the Biichel et al. (2015) cited in the
LSHTM protocol, corrected the data to reflect endpoints determined according to
EPA's guidelines (first crossing vs. first confirmed crossing) and analyzed those
data to determine the P5MR for the three repellents tested. The P5MRs ranged from
0.269 for EBAAP to 0.539 for Icaridin. The P5MRs for DEET products fell within
that range.
• Given the small sample size (10 subjects), EPA did not find it appropriate to
attempt to establish a confidence interval around these P5MR values.
• For EPA to accept a study as valid for showing a product's efficacy, the P5MR
should be greater than or equal to 0.4. Study results showing P5MRs of 0.3-0.4 will
generally be considered on a case by case basis. Any results where the P5MR is
lower than 0.3 are unlikely to meet EPA's criteria for efficacy of a repellent
product. The P5MR value indicates the "peakedness" of the CPT distributions (or
the distributions of point of efficacy failure), with higher P5MRs indicating that
more of the distribution is closer the median CPT. It is a measure of the "spread" of
distribution of CPT for a given product, with flatter distributions suggesting greater
expected variability in CPTs among users of the product. The lower the P5MR, the
higher the spread or variation in CPT, and the greater the proportion of the
population of users likely to experience protection times substantially less indicated
on the product label. The protection time on the product label is generally the
median CPT, and a small P5MR value indicates that a substantial portion of the
population will realize a significant departure from that labelled protection time, or
the median CPT.
• Based on the results of EPA's simulations, where the desired K=0.7, the median
CPT is 4 or 6 hours, and where the expected P5MR is 0.4, the number of subjects
likely to achieve sufficient precision with adequate power of at least 80% precision
is 23-28 (Table A5).
• EPA (science, statistics, and ethics) and LSHTM agreed that a sample size of 25
would be adequate to ensure that the study includes enough subjects to return
reliable results without unnecessarily including more subjects than necessary.
Page 14 of 70
-------
In addition, the study protocol proposes to pool all the data from all tick species tested
on all volunteers to calculate period of lasting repellency, defined as to the time point at
which less than 90% of the ticks are repelled. "The data from all tick [species] on all
volunteers will be pooled and the protective efficacy will be calculated for each time
point using the formula below. The time at which protective efficacy fall below 90% will
be reported.
Protective efficacy = (Total ticks tested - Total ticks not revelled) x 100
Total ticks tested"
(pg. 18, Section 9.2 Statistical methods). However, it is recommended that this
objective be removed from the protocol.
10. Withdrawal criteria:
Participants can withdraw at any time without giving a reason for withdrawing. Data
collected to the point of withdrawal will be used in the statistical analysis of the data
"unless the participant requests that their data is not used, in which case it will be
removed from the database. Participants may also be removed at the discretion of the
Chief Investigator, where continued participation may affect the safety of the participant
or where there is a development of any condition which might interfere with study
participation " (pg. 13, section 5.4. Withdrawal Criteria).
11. How and to what will human subjects be exposed/ Product description:
The products, are registered products (EPA Reg. No. 84878-2, 305-62), containing the
active ingredient Citriodiol, also known as Oil of Lemon Eucalyptus (OLE). Citriodiol®
contains a minimum of 65% /;-Menthane-3,8-diol (PMD), which is the active
component in OLE, as well as other naturally occurring constituents present in the
essential lemon eucalyptus oil from which it is derived. Copies of submitted Master
label include Sublabel A for liquid pump spray packaging, and Sublabel B for
pressurized (bag-on-valve) (BOV) packaging.
Subjects will be exposed to the repellent product, EPA Reg. No. 84878-2 or EPA Reg.
No. 305-62 for 10 hours of testing against each of 3 species of pathogen-free adult ticks
from laboratory-reared tick colonies. Proposed exposure periods consist of exposing
ticks to untreated human skin for screening, and exposing 1 individual tick to treated
forearms for 3 minutes, at intervals of 15 minutes for a maximum of 10 hours of testing
per species, or until the time point when repellent breakdown or CPT is reached by
subject, whatever happens sooner.
EPA Reg. Nos. 84878-2, and 305-62, containing 30 % OLE, are categorized as Toxicity
Category III for acute oral toxicity (LD50 > 2,000 mg/kg), acute dermal toxicity (LD50 >
2,000 mg/kg), and eye irritation (moderate irritant), and toxicity category IV for dermal
irritation. OLE is not a skin sensitizer.
Page 15 of 70
-------
The risk assessment for OLE is based on the EPA risk assessment for p-Menthane-3,8-
diol (PMD), which is the active component in OLE. Citriodiol contains 65% PMD
according to EPA's risk assessment (1999). A 90-Day dermal study in rats (MRID
444387-10) tested PMD (98.3 % pure) at increasing doses, 0, 1,000 and 3,000
mg/kg/day. The NOAEL = 1,000 mg/kg/day, and the LOAEL = 3,000 mg/kg/day. The
endpoints for NOAEL and LOAEL are based on treated skin observations, erythema,
edema, eschar, and histological observations in treated skin, increased acanthosis and
inflammation at the highest dose of 3,000 mg/kg/day. No dermal absorption data are
required for Tier I Toxicity data for registration of biochemical products therefore,
without these data, dermal absorption is assumed to be 100%. Risk characterization for
infants and children is based on data from one developmental study (MRID 444387-11)
in which the NOAEL =3,000 mg/kg/day. No LOAEL was established, and thus, a 10-
fold safety factor is applied for risk characterization. MOEs were not calculated because
there are no endpoints of concern for the dermal route of exposure. The Agency
concluded that there is reasonable certainty of no harm to populations or subpopulation
(infants and children) from the use of PMD in insect repellent products applied to
human skin.
12. GLP and OA
"This study will adhere to the principles outlined in the International Conference on
Harmonization Good Clinical Practice (ICH GCP) guidelines, Good Laboratory
Practices (as defined by 40 CFR part 160), protocol and all applicable local
regulations." (pg. 2).
Arctec will assign the independent Quality Assurance Unit (QAU) person, who is
independent from the study sponsor and will perform all QA duties.
13. Good Laboratory Practice (GLP) Compliance and Quality Assurance
"Good Laboratory Practices, as defined by 40 CFR part 160, will be followed
throughout this study, arctec will be a member of the UK GLP compliance monitoring
programme (GLP-CMP) before any trial procedures are undertaken. The UK is part of
the OECD GLP Mutual Acceptance of Data agreement with the USEPA. Under this
agreement, UK laboratories that are members of the UK GLP-CMP have met the US
GLP burden of compliance, and data produced as such will be accepted as would data
generated in the US under the USEPA GLP regulations. The UK GLP-CMP is run by
the UK GLP Monitoring Authority (UK GLPMA). The GLPMA will inspect the test
facility every 12-30 months, including an audit of completed and on-going studies to
check that the principles of GLP have been complied with. Under the UK GLPMA, the
QA unit will be a person employed by arctec, who is independent of study conduct and
directly responsibly to management. The QA representative will conduct critical phase
inspections at intervals adequate to ensure study integrity, and maintain written and
Page 16 of 70
-------
signed records of each inspection. Records shall identify the study and include the date
of the inspection, the phase inspected, the individual conducting the inspection, positive
and negative findings, actions recommended and taken to resolve negative findings, the
scheduled date for re-inspection (if any), and the date(s) the findings are reported. All
inspection findings will be reported to management and the Study Director. Any
problems, amendments or deviations discovered shall be brought to the attention of the
sponsor, Study Director and management immediately. The QA representative will
review the final reports for accuracy and compliance with GLPs and the protocol. A
signed QA statement will be included in the final report that lists the phase inspections
that were conducted, their dates, and the dates the findings were reported to
management and the Study Director. Under UK GLPMA, the responsibility for
producing a statement of compliance rests with the Study Director. " (pg. 8, Section 3).
14. Compliance with EPA. FIFRA (Federal Insecticide. Fungicide and Rodenticide
Act), and Good Laboratory Practice Standards (GLP); 40 CFR. Part 160 (October
1989):
The UK is part of the OECD GLP Mutual Acceptance of Data agreement with the
USEPA. Under this agreement, UK laboratories that are members of the UK GLP-CMP
have met the US GLP burden of compliance, and data produced as such will be accepted
as would data generated in the US under the USEPA GLP regulations. Arctec will be a
member of the UK GLP compliance monitoring programme (GLP-CMP). The UK
GLP-CMP is run by the UK GLP Monitoring Authority (UK GLPMA). Under the UK
GLPMA, the QA unit will be a person employed by arctec, who is independent of study
conduct and directly responsibly to management. The QA representative will conduct
critical phase inspections at intervals adequate to ensure study integrity, and maintain
written and signed records of each inspection.
Study site location and testing facility: Arthropod Control Product Test Center (arctec),
a division of Chariot Innovations Limited. A wholly-owned subsidiary of the London
School of Hygiene & Tropical Medicine, (pg. 4)
Study Director: Professor James Logan, arctec, LSHTM, Chief Investigator (pg. 3)
Study Sponsor: Citrefine International Ltd. Moorfield Rd., Yeadon, Leeds. LS19 78N,
UK
E. Compliance with Applicable Scientific Standards
This protocol adequately addresses the following elements according to applicable
scientific standards:
• Experimental design
• Data analysis
• Risk minimization
Page 17 of 70
-------
F. EPA Science Comments
The study sponsor has agreed to make changes per the recommendations below with the
exception of using the standard dose of 0.5 mg product/600 cm2. EPA's recommendation
and rationale for using the standard dose is presented in the science comments below.
Instead of the standard dose, the sponsor intends to conduct a dosimetry phase to determine a
typical consumer dose.
In summary, the following sections in the study protocol should be revised according to the
following recommendations before the research goes forward:
1. Objectives: Determination of 90% repellency duration should be removed from study
objectives in Study Synopsis, pg.3, and in Section l.a. Objectives, pg. 7, and from Section
9.2 Statistical methods, pg. 18, in the protocol. The objective of the proposed study is to
estimate lasting efficacy of the repellent product, expressed as complete protection time.
This measure of efficacy is defined as the period of time between product application and
first confirmed crossing. The first confirmed crossing signals repellency failure or CPT. First
confirmed crossing is the endpoint used to measure lasting efficacy. Therefore, to be
consistent, lasting efficacy doesn't need to be measured, or expressed, by sustained 90%
repellency. In addition, update this section of the protocol by adding product EPA Reg. No.
305-62 to the objectives, and revise EPA Reg. No. 84878-2 (alt). Identify "(alt)," and
explicitly state the registration numbers of the products proposed for testing throughout the
document.
2. Endpoints: The definition of endpoints should be revised. The endpoint for estimation of
CPT should be identified as the time to first confirmed crossing, signaling the time point of
repellency failure or complete protection time.
Efficacy endpoint is defined on pg. 4, Study Synopsis; on pg. 9, Section 4.1 Trial Endpoints,
and on pg. 18, Section 9.2. Statistical methods, as complete protection time, which is a
measure of lasting repellency. The efficacy endpoint however, is the First Confirmed
Crossing, which is defined in Section 9.2. Statistical methods, as 1 tick crossing 3 cm of
treated skin, followed by a second confirmatory crossing [by another tick] within 30 minutes
apart. The CPT used for the product should be the tick species which results in the lowest
CPT.
Dosimetry endpoint, or dose determination is defined in Study Synopsis, pg. 4, and on pg. 9,
Section 4.1 Trial Endpoints. It is defined in Section 9.1. Statistical methods, pg. 18, as the
mean dose from a sample of 21 volunteers, measured as mg/cm2 and converted to volume
using the specific gravity value of the formulation. Endpoint of dosimetry should be
accurately defined throughout the document consistently with the definition given in Section
9.1. Statistical methods, pg. 18, "as the mean dose from a sample of 21 volunteers, measured
as mg/cm2 and converted to volume using the specific gravity value of the formulation." In
addition, update sample size to 25 subjects.
Page 18 of 70
-------
3. Sample size: LSHTM has agreed with EPA that a sample size of 25 would be adequate to
ensure that the study includes enough subjects to return reliable results without including
more subjects than necessary (Attachment 3 of this review; Table A5). Thus, the protocol
should be updated to a sample size of 25 test subjects for efficacy testing, and the number of
alternate subjects in section 6.2. Randomization, pg. 13, should be increased accordingly
(See item #4. Number of alternate participants). In addition, using the same number of
subjects (25) is also recommended for dosimetry testing.
4. Number of alternate participants: Update Section 6.2. Randomization, on pg. 13 of the
protocol to state the number of alternate subjects, including distribution of their sex ratio
according the amended sample size of 25 subjects. The total number of participants (75
participants) in section 6.2. Randomization, pg. 13, includes alternates, that number should
be updated according to amended sample size of 25 test subjects per tick species. In
addition, consider recruiting alternate subjects during recruitment process, and describe how
participants will be randomly assigned as either test subjects or alternates.
5. Estimation of skin surface area: Describe in more detail how to measure the forearm
surface area. For example, the narrative could include the use of 4 or 3 equidistant bracelets
(i.e., 4 or 6 cm wide) placed above the wrist to determine the average circumference of the
forearm at more than 2 points, wrist and elbow as proposed, and multiply the average
circumference by the length of the forearm. Wrist and elbow are not enough to estimate the
average circumference of the forearm. (OCSPP 810.3700 Guidelines, (i) Specific guidance
for Dose Determination, (3) Measuring subject's skin area). "The surface area of subject's
limb in cm2 can be estimated by measuring the circumference of the forearm in centimeters
at the wrist and elbow, or of the leg at the ankle and knee, and in either case at one or two
equally spaced intermediate points"
6. Dose application and time to first exposure: The statement"Participants will be asked to
avoid rubbing or washing the arm for the first 20 minutes after application " in Section 6.
Interventions, pg. 13, is unclear. Explain why participants are not asked not to disrupt treated
skin throughout the entire period of testing. In addition, clarify whether the 20 minutes after
application (as stated in Section 6. Interventions, pg. 13), indicates that testing (exposure of
ticks to treated skin) will begin 20 minutes post-application. It is not clear whether exposure
of ticks to treated skin will start 20 minutes post-application, or whether exposure of ticks to
treated skin will begin immediately after application.
7. Definition of "crossing" and criteria for repellency: Define "crossing" as stated in the
OCSPP 810.3700 Guideline, (b) Definitions (i) (iv) "A crossing is the act of passage by a
tick or chigger from an area of untreated skin to an area of treated skin. A crossing may be
quantified either or both by the distance the tick or chigger moves onto treated skin or by
how long the tick or chigger remains on treated skin. " Add the criteria for repellency to
Section 8.2. Assessment of complete protection time on pg. 16 of the study protocol. If the
crossing line is eliminated from the design then, the criteria for no repellency should be
Page 19 of 70
-------
amended as "a tick is classified as not repelled when it crosses the boundary line and spends
1 minute on treated skin." In addition, describe the criteria for determination of first
confirmed crossing: Breakdown of the product occurs when a tick crossing unto treated skin
is not repelled, and it is followed by a second tick that is also not repelled, or it is followed by
a second tick that is repelled, but if the second tick is followed by the next tick (the third
tick), and that third tick is not repelled then, the third tick is considered a confirmatory
crossing within a 30 minute period for the first tick that was repelled.
8. Statistical analysis and sample size calculation: EPA (science, statistics, and ethics) and
LSHTM agreed that a sample size of 25 would be adequate to ensure that the efficacy study
includes enough subjects to return reliable results without including more subjects than
necessary. The same sample size and the same subjects should also be employed for
dosimetry.
9. Alternate subjects and subjects' withdrawal: Specify that alternate subjects will replace
individual subjects who withdraw from the study. If a subject withdraws after a full day of
testing with one tick species, they will be replaced for testing with the next tick species. The
data for the completed test day will be used. However, if subjects withdraw before
completing a test day their data will not be used and they will be replaced with an alternate
subject.
10. Standard versus consumer dose: Standard dose (EPA recommendation): Although
EPA's product performance guidance includes steps for a dose determination phase as part of
an insect repellent efficacy study, to minimize the exposure of test subjects, EPA
recommends using a standard dose based on previously reviewed and accepted studies.
Based on an analysis of dosimetry results from repellent studies reviewed by EPA and HSRB
since 2006 (Table 1), EPA considers the dose of 1 g product/600 cm2 of skin to be an
appropriate product dose for testing aerosol, wipes, and lotion type products and 0.5 g (±
10%) product/600 cm2 of skin for testing pump spray type products under this protocol.
Table 1. Combined results of dosimetry testing from skin-applied repellent studies reviewed
by EPA and HSRB since 2006 for three formulation types.
Formulation Total No. of Subjects in Mean Dose
Type Dosimetry Phase for Dose (g/600 cm2) range
Mosquito Tests + 1 SD (g/600
cm2)
Lotion 112 0.933 +0.299 0.63-1.23
Pump spray 102 0.417 + 0.120 0.28-0.56
Aerosol 25 0.815 + 0.262 0.55-1.08
Page 20 of 70
-------
Reasons for using a set standard application rate rather than Dosimetry:
1. Minimize variability. Variable application rates between studies make it impossible to
determine whether performance of similar formulations (pump, lotions or aerosols)
tested at different consumer doses is driven by variable application rates or by the test
products per se.
2. Dosimetry testing introduces additional test subjects for an adequate sample size.
3. The standard application rates for pump spray products are based on dosimetry data
from 10 repellent studies previously reviewed by EPA/HSRB (Table 1). One of these
studies was conducted using EPA Reg. No. 305-62, one of the products proposed for
testing. In the dosimetry phase for EPA Reg. No. 305-62, the average typical dose
applied to subject's legs (n = 10) was 0.28 g/600 cm2. Although the registrant
submitted a rationale indicating that pump sprays will be applied at higher rates to
subject's arms, EPA's analysis contains five studies conducted with pump sprays
applied to subject's arms and the mean typical consumer dose (± SD) from only these
six studies is 0.496 ± 0.09 g product/600 cm2. Also, as seen in Table 1 the high end of
the dose range for pump sprays is 0.56 g/600 cm2.
Therefore, EPA expects the results of a dosimetry phase conducted with the proposed
products to be consistent with the results of previous studies.
Dosimetry phase (Sponsor preference): The study sponsor expects the typical consumer dose
rate to be substantially higher than EPA's recommended standard dose, and therefore
believes the recommended standard dose will result in an underestimate of the actual
protection time against ticks offered by this product for the following reasons.
1. The data relied on to support EPA's standard dose as based in whole or in part upon
dose rate studies that only measure self-application to the lower leg. The relevance of
these data to this tick study is questionable because this study will assess application
to the consumer's forearm.
2. The pump spray mechanism functions optimally when used upright or at a slight
angle, and flow-through is reduced when consumers angle the pump spray
mechanism significantly as naturally occurs to apply to areas below the waist.
3. Risk associated with exposure of human subjects during a dosimetry phase de
minimus; the product is already registered by EPA and has been on the market for
over a decade. Additionally, in this study the product is only being applied to a
fraction of the skin area to which a consumer would be expected to apply.
10. Calculation of consumer dose: If dosimetry testing is preferred, the protocol should
describe the method employed for quantification of applied product, including how the
average consumer dose is calculated. For example, the study protocol should show the
calculations that will be employed for quantification of applied product, the mean dose
applied by each subject to each limb, and the grand mean across all subjects' means. (OCSPP
810.3700 Guidelines, (i) Specific Guidance for Dose Determination (5) Calculating standard
dose for use in repellency trials).
Page 21 of 70
-------
11. Remove the reference to previous consumer dose testing with this product because EPA
cannot rely on those data.
12. Mode of application for determination of consumer dose: The statement on pg. 15,
Section 8. Test Methodology, Subsection 8.1. Consumer dose determination, "Participants
[for dosimetry testing] will be asked to apply the repellent to their forearm only as if they
were about to enter a tick-endemic area'' should be removed from the protocol. Subjects
should be instructed to follow label recommendations for application of the product to their
forearms.
13. Please include copies of the data collection sheets as part of the study protocol in a
separate Appendix.
14. On pg. 29, §14.2 Sample size calculation, it is not explained how the level of precision
for a 95% confidence interval of 0.1 |il/cm2, e.g., example 1.5-1.6 |il/cm2, was calculated
(Table 2 below corresponds to Table A1 under the same section) to determine sample size of
21 subjects. Consider omitting this part in the protocol. This sample size calculation could
be eliminated if the same sample size for 25 subjects already determined for efficacy testing
is also employed for dosimetry testing.
Table 2. Sample size calculations to determine the consumer dose with a specified level of
precision [0.1 |il/cm2] (reproduced from Table Al, pg. 29 of the Study protocol).
Size of 95%
Uncorrected
Cluster size B
Number of
confidence
sample size A
participants
interval
0.4
4
9
2
0.3
7
9
3
0.2
16
9
6
0.1
61
9
21
0.075
109
9
37
0.05
244
9
82
' according to the publication of Kirkwood and Stern (2003)
1 according to the publication of Hayes and Bennet (1999)
15. Add the following to section 8.1 Consumer Dose Determination on pg. 15: "the typical
consumer dose in mg/cm2 will be converted to volume using the specific gravity of the test
material."
16. On pg. 15, § 8.2, state that control arm will be arranged in an identical fashion as treated
arm, but without repellent. In addition, questing behavior on the control arm should be
described as "a tick that moves steadily from the release line across boundary line and
upward."
Page 22 of 70
-------
17. On pg. 16, § 8.2, the statement: "This will be repeated so that five actively questing ticks
will be exposed to the treated arm one at a time, at 30 minute intervals (timedfrom product
application) for 10 hours or until treatment failure" - should be changed if the exposure
model is changed to 1 tick every 15 minutes. Support this change with the following
rationale:
"EPA revised the model discussed with LSHTM in December to include simulations for up
to 30 subjects doing testing with 5 ticks every 30 minutes. EPA also developed simulations
for up to 30 subjects doing testing with 1 tick every 15 minutes, another option for testing
listed in EPA's guidelines (https://www.regulations.gov/document?D=EPA-HQ-QPPT-2009-
0150-0011)."
18. Formulation of Products on pg. 13, § 6, subsection 6. Information about product
composition is considered confidential business information and should be removed from the
protocol.
19. Risks and Benefits - In section 4.2, make the statement clear that for individual subjects,
at least 48 hours will lapse between test days involving exposure to ticks.
20. Please make the following corrections on the data sheets collection provided to EPA:
a) For dosimetry phase, change the directions for the practice applications to read as
follows "Give the participant the product bottle and ask them to read the label
instructions and then practice spraying the product onto their forearm. DO NOT offer
advice on the application method."
b) When calculating the consumer dose rate on the forearm measurement & dose
calculation sheet, report the amount in weight units.
c) In the section on preparation of participant's arms, make sure that release and
boundary lines reference the correct locations (e.g., boundary line will be drawn at the
wrist, and release line will be located 3 cm below boundary line.)
Attachments:
1. EPA Protocol Review (Protocol dated July 21,2017)
2. Completeness checklists
3. EPA's Analysis & Support for Sample Size
4. EPA's comments on the IRB-approved protocol
5. EPA's comments on the IRB-approved informed consent materials
6. IRB approval of protocol dated July 21, 2017
7. IRB approved protocol & informed consent materials
8. IRB correspondence
9. IRB minutes
Page 23 of 70
-------
Page 24 of 70
-------
Attachment 1 - EPA Protocol Review
Title: A single group trial to determine the complete protection time of an insect repellent
formulation containing 30% Citriodiol® (Oil of Lemon Eucalyptus) against three species
of ticks
Date: July 21, 2017
Principal Investigator and any sub-investigators: Dr. James Logan, arctec
Participating Laboratory: Arthropod Control Product Test Centre {arctec), a division of
Chariot Innovations Limited, a wholly-owned subsidiary of the London School of Hygiene &
Tropical Medicine (LSHTM).
Sponsor: Citrefine International Ltd. Moorfield Rd., Yeadon, Leeds. LSI9 78N, UK
Trial Monitoring Center:
arctec
Room LG38, LSHTM
Chariot Innovations Ltd.
Keppel St.
London WC1E7HT
UK
IRB:
Western Institutional Review Board
1019 39th Avenue SE Suite 120
Puyallup, WA 98372-2115
1. Societal Value of Proposed Research
(a) What is the stated purpose of the proposed research?
This study is designed to determine the complete protection time (CPT) of EPA
registered repellents, EPA Reg. No. 84878-2 and 305-62, containing 30% Citriodiol®
(Oil of Lemon Eucalyptus (OLE)) as its active ingredient, against 3 adult tick species,
(pathogen free) Amblyomma americanum, Rhipicephalus sanguineus, and Ixodes
scapularis, at an average consumer dose. The product will be tested on human subjects
in a laboratory setting for up to 10 hours. EPA requires efficacy testing of products
claiming efficacy against disease vectors to support efficacy claims on product labels.
"The primary objectives are to determine the complete protection times of the
formulation corresponding with EPA Reg. No. 84878-2 (alt) insect repellent against
adult Ixodes scapularis, Amblyomma americanum and Rhipicephalus sanguineus applied
at the consumer dose rate " Study (Synopsis, pg. 3). The aim of the proposed study is to
determine the duration of efficacy of an insect repellent containing 30% OLE against
three tick species. The secondary objective is to determine the average consumer dose
for assessment of efficacy. (Section 1. a. Objectives (pg. 7).
Page 25 of 70
-------
(b) What research question does it address? Why is this question important?
Would the research fill an important gap in understanding?
The purpose of the study is to determine the mCPT of a personal, skin-applied tick repellent
product, containing the active ingredient OLE. This information does not currently exist.
The rationale for testing is to collect data to establish a median complete protection time.
The data supporting currently registered products are not sufficient to establish a median
complete protection time for the product discussed in the protocol.
A standardized protocol will enable the EPA to receive consistent and scientifically reliable
data about the complete protection time for the product. The laboratory testing data will
provide information about: 1) a "consumer dose" derived from 25 subjects each making 3
applications of the product to one forearm, and 2) the length of time after treatment before
the first confirmed crossing by a tick. Ticks will not be permitted to bite subjects. The
endpoint is crossing, and ticks attempting to attach will be removed before they can bite a
subject.
(c) How would the study be used by EPA?
EPA requires product-specific efficacy data conducted to assess skin applied insect
repellent products in terms of the recommendations of the EPA OPPTS 810.3700
Guideline. These products have not been evaluated for their performance against ticks.
EPA will review the study to satisfy product specific efficacy data requirements and
acceptable label claims for repellent efficacy for the test products.
(d) Could the research question be answered with existing data? If so, how? If not,
why not?
EPA requires product-specific efficacy data to support product registration. No previous
testing of this product against ticks under the proposed use pattern has been conducted.
(e) Could the question be answered without newly exposing human subjects? If so,
how? If not, why not?
Human subjects are required because they represent the target system for the test
material, and sufficiently reliable non-human models for repellency testing have not been
developed.
2. Study Design
(a) What is the scientific objective of the study? If there is an explicit hypothesis, what
is it?
"The aim of this study is to determine the duration of efficacy of an insect repellent
containing 30% OLE against three tick species.
The primary objectives are to:
Page 26 of 70
-------
1. Determine the complete protection time (CPT) of EPA Reg. No. 84878-2 and 305-
62 formulated insect repellents against adult Ixodes scapularis applied at the
consumer dose rate.
2. Determine the CPT of EPA Reg. No. 84878-2 and 305-62 formulated insect
repellents against adult Amblyomma americanum applied at the consumer dose
rate.
3. Determine the CPT of EPA Reg. No. 84878-2 and 305-62 formulated insect
repellents against adult Rhiyiceyhalus sanguineus applied at the consumer dose
rate.
The EPA has established consumer dose rates for pump sprays of 0.83 mg/cm2. As these
two figures are so different, a consumer dose rate study would allow this protocol to test
a realistic dose rate, rather than relying on rates established for other products. The
secondary objectives are to: 1) determine the average dose rate applied by consumers
when using EPA Reg. No. 84878-2) formulated insect repellent to repel ticks" (pg.7, §2).
(b) Can the study as proposed achieve that objective or test this hypothesis?
The objective cited may be achieved by the study if the protocol is revised and amended
in accordance with EPA's comments on the ethical and scientific aspects of the protocol.
2.1 Statistical Design
(a) What is the rationale for the choice of sample size?
The original protocol submitted arctec proposed that 10 individuals serve as test
subjects. However, the justification for the proposed sample size is not supported
statistically. After consultation with EPA, arctec and the Study Sponsor have agreed
to use EPA's recommended sample size of 25 test subjects, the methodology for
which is described in Attachment 3 to EPA's science and ethics review of the
protocol. A sample size of 25 would be adequate to ensure that the study includes
enough subjects to return reliable results without unnecessarily including more
subjects than necessary (pg. 18, § 9).
(b) What negative and positive controls are proposed? Are proposed controls
appropriate for the study design and statistical analysis plan?
One arm from each subject will remain untreated and serve as the untreated control
for the purpose of screening active questing ticks, and the other arm will be treated
with the test product and serve as the treatment group. A positive control will not be
used (pg.16, § 8.2).
(c) How is the study blinded?
The study is not blinded. There is only one product tested at a time, and observations
are based on tick behavior.
(d) What is the plan for allocating individuals to treatment or control groups?
Page 27 of 70
-------
"The treatment will be applied to either the left or the right arm of the participants
depending on a randomization schedule to be produced by the trial statistician. The
schedule will assign treatment to either the left or right forearm for each of the three
tick species for each participant up to eighty participants. This number should be
sufficient to cover the participant numbers of volunteers who consent, but fail
screening or who withdraw or are withdrawn before taking part in efficacy testing"
(pg- 13, § 6.2).
(e) Can the data be statistically analyzed?
Yes. See (f) below.
(f) What is the plan for statistical analysis of the data?
The median CPT of all test subjects for each tick species will be calculated using the
Kaplan-Meier survival analysis. The duration of protection for the repellent product
will be the lowest median CPT of the 3 tick species (pg. 18, §9). EPA recommends
rounding down to the nearest whole number. For example, three hours and 45
minutes would be listed on the label as three hours
(g) Are proposed statistical methods appropriate to answer the research question?
The median CPT, will be estimated from the CPT for each participant per tick
species, using Kaplan-Meier survival analysis. The Kaplan Meier procedure is a non-
parametric method for survival analysis; this method does not require or assume the
data to follow a particular parametric distribution. This method can also account for
censored observations. Kaplan-Meier estimator has been accepted by EPA and the
HSRB for median CPT calculation in past repellent efficacy studies and is also
recommended by the World Health Organization for CPT calculation from these non-
parametric data sets.
(h) Does the proposed design have adequate statistical power to definitively answer
the research question?
The sample size has been revised and updated according to EPA recommendations for a
sample size of 25 subjects. This recommendation is based on the results from EPA's
simulations, where the desired K=0.7, the median CPT is 4 or 6 hours, and where the
expected P5MR is 0.4, the number of subjects likely to achieve sufficient precision with
adequate power of at least 80% precision is 23-28. For EPA to accept a study as valid for
showing a product's efficacy, the P5MR should be greater than or equal to 0.4. Study
results showing P5MRs of 0.3-0.4 will generally be considered on a case by case basis.
Any results where the P5MR is lower than 0.3 are unlikely to meet EPA's criteria for the
efficacy of a repellent product. The P5MR value indicates the "peakedness" of the CPT
distributions (or the distributions of point of efficacy failure), with higher P5MRs
indicating that more of the distribution is closer to the median CPT. It is a measure of the
"spread" of distribution of CPT for a given product, with flatter distributions suggesting
Page 28 of 70
-------
greater expected variability in CPTs among users of the product. The lower the P5MR,
the higher the spread or variation in CPT, and the greater the proportion of the population
of users likely to experience protection times substantially less indicated on the product
label. The protection time on the product label is generally the median CPT, and a small
P5MR value indicates that a substantial portion of the population will realize a significant
departure from that labelled protection time, or the median CPT. For detailed information
on the statistical simulation see Attachment 3 of this review.
2.2 How and to what will human subjects be exposed?
Subjects will be exposed to an EPA-registered repellent product, either EPA Reg. Nos.
84878-2 and 305-62 containing 30% oil of lemon Eucalyptus (OLE). The active
ingredient in OLE is 65 % p-Methane-3,8-diol (PMD). The application of the product to
the skin of subjects in the proposed study will be consistent with the label directions for
use. The active and inert ingredients of a currently registered formulation have
undergone EPA review and the requirements are fulfilled for EPA registration of
repellent product.
Subjects will be exposed to ticks during the laboratory-based repellent testing. To
eliminate the risk of transmission of tick-borne disease, ticks will be sourced from
pathogen-free colonies at Oklahoma State University Tick Rearing Facility. Ticks will
only be used once with a single subject, and ticks will be destroyed by immersion in
alcohol after they are placed on a subject's arm.
(a) What is the rationale for the choice of test material and formulation?
Efficacy data to satisfy product performance requirements and to support label claims
of repellency against ticks for this product are required by EPA for registration. EPA
requires submission of product performance data for all products claiming efficacy
against public health pests.
(b) What is the rationale for the choice of dose/exposure levels and the staging of
dose administration?
The rationale for testing is to collect data to establish a median CPT for a product
containing OLE. The data supporting currently registered OLE products do not
provide this information.
The dose for the repellency phase of the study will be the grand mean of the 3 doses
applied by each of the 25 subjects during the dosimetry phase.
Standard dose (EPA recommendation): Although EPA's product performance
guidance includes steps for a dose determination phase as part of an insect repellent
efficacy study, to minimize the exposure of test subjects, EPA recommends using a
standard dose based on previously reviewed and accepted studies. Based on an
analysis of dosimetry results from repellent studies reviewed by EPA and HSRB
since 2006 (Table 1), EPA considers the dose of 1 g product/600 cm2 of skin to be an
Page 29 of 70
-------
appropriate product dose for testing aerosol, wipes, and lotion type products and 0.5 g
(± 10%) product/600 cm2 of skin for testing pump spray type products under this
protocol.
Table 1. Combined results of dosimetry testing from skin-applied repellent studies
reviewed by EPA and HSRB since 2006 for three formulation types.
Formulation Total No. of Subjects in Mean Dose range
Type Dosimetry Phase for Dose (g/600 cm2) (g/600 cm2)
Mosquito Tests + 1 SD
Lotion 112 0.933 + 0.299 0.63-1.23
Pump spray 102 0.417 + 0.120 0.28-0.56
Aerosol 25 0.815 + 0.262 0.55-1.08
Reasons EPA recommends using a set standard application rate rather than dosimetry:
1. Minimize variability. Variable application rates between studies make it
impossible to determine whether performance of similar formulations (pump,
lotions or aerosols) tested at different consumer doses is driven by variable
application rates or by the test products per se.
2. Dosimetry testing introduces additional test subjects for an adequate sample size.
3. The standard application rates for pump spray products are based on dosimetry
data from 10 repellent studies previously reviewed by EPA/HSRB (Table 1). One
of these studies was conducted using EPA Reg. No. 305-62, one of the products
proposed for testing. In the dosimetry phase for EPA Reg. No. 305-62, the
average typical dose applied to subject's legs (n = 10) was 0.28 g/600 cm2.
Although the registrant submitted a rationale indicating that pump sprays will be
applied at higher rates to subject's arms, EPA's analysis contains five studies
conducted with pump sprays applied to subject's arms and the mean typical
consumer dose (± SD) from only these six studies is 0.496 ± 0.09 g product/600
cm2. Also, as seen in Table 1 the high end of the dose range for pump sprays is
0.56 g/600 cm2.
Dosimetry Phase (Sponsor Preference): The study sponsor expects the typical
consumer dose rate to be substantially higher than EPA's recommended standard
dose. The Study Sponsor proposes to conduct a dosimetry phase to derive a realistic
consumer dose for this product, which impact the repellent testing. The Study
Sponsor expects that the dose derived from the dosimetry phase would result in an
accurate CPT for this product, thereby preventing a significant underestimate of the
efficacy, which in turn could result in millions of consumers over-applying the
product. The Study Sponsor believes using EPA's recommended standard dose
Page 30 of 70
-------
would result in an underestimate of the actual protection time against ticks offered by
this product for the following reasons.
1. The data relied on to support EPA's standard dose as based in whole or in part
upon dose rate studies that only measure self-application to the lower leg. The
relevance of these data to this tick study is questionable because this study will
assess application to the consumer's forearm.
2. The pump spray mechanism functions optimally when used upright or at a slight
angle, and flow-through is reduced when consumers angle the pump spray
mechanism significantly as naturally occurs to apply to areas below the waist.
3. Risk associated with exposure of human subjects during a dosimetry phase de
minimus; the product is already registered by EPA and has been on the market for
over a decade. Additionally, in this study the product is only being applied to a
fraction of the skin area to which a consumer would be expected to apply.
(c) What duration of exposure is proposed?
The exposure is a 3-minute period at 15 minute intervals. The tick is placed on the
untreated arm to determine if it is attracted to the subject's skin. After a tick is
deemed to be attracted to the subject, and actively questioning, the tick is then placed
on the treated arm for 3 minutes. This exposure process is repeated every 15 minutes
with a different questing tick, and thereafter for 10 hours or until the test substance
fails to repel the tick (first confirmed crossing takes occurs), (pg. 16, § 8.2).
2.3 Endpoints and Measures
(a) What endpoints will be measured? Are they appropriate to the question(s) being
asked?
Efficacy endpoints are defined on pg. 4, Study Synopsis, as "Time to complete
protection time measured as a single tick crossing 3cm into treated skin, and time to
90 % protective efficacy measured as more than 10% of ticks crossing treated skin.,,
Efficacy endpoints are defined on pg. 9, Section 4.1 Trial Endpoints, as "Theprimary
endpoint measure is the complete protection time (CPT), which is defined as the time
between application of the repellent product and the time at which 1 tick crosses 3cm
of treated skin, followed within 30 minutes by a second confirmatory crossing. The
median CPT will be calculated for each species using Kaplan-Meier survival curves.
The first secondary endpoint measure is the averase dose rate applied by consumers
when using EPA Reg. No. 84878-2formulated insect repellent according to label
instructions for purposes of repelling ticks. Other secondary endpoint measures will
be the time to 90% protective efficacy of EPA Reg. No. 84878-2formulation against
Ixodes scayularis. Amblyomma americanum and Rhiyiceyhalus sanguineus."
The endpoint for estimation of CPT is correctly defined on pg. 18, Section 9.2.
Statistical methods, as the time to failure, which is the time when a first tick crossing
into 3 cm of treated skin is followed by a second confirmatory crossing from a
different tick within 30 minutes from the first crossing. "The endpoint for CPT will
be time to treatment failure for each participant test. Treatment failure is the time at
which the product no longer provides complete protection, which is determined as the
Page 31 of 70
-------
time at which 1 tick crosses 3 cm of treated skin, followed within 30 minutes by a
second confirmatory crossing. " (pg. 18, § 9.2. Statistical methods. If the crossing
line is removed from the design, the criteria for repellency is changed to a tick that
crosses the boundary line and spends 1 minute on treated skin.
The endpoint for dosimetry is defined in the same section as "The typical consumer
dose of one insect repellent product measured as ml/cm2. " And It is also defined on
pg. 18, Section 9.1. Statistical methods as "The endpoint for the consumer dose
assessment is the mean dose from all 21 volunteers in mg/cm2. This will be converted
to ml/cm2 taking into account the specific gravity of the test material. "
Using the Kaplan-Meier estimator, the Median CPT will be calculated for all test
subjects exposed to each tick species.
The endpoints are appropriate to the questions being asked.
(b) What steps are proposed to ensure measurements are accurate and reliable?
• Good Laboratory Practices, as defined by 40 CFR part 160 will be followed
throughout all studies.
• Study staff will prepare forearm for dose application, and will measure skin
surface area in advance, and perform dose application for testing.
• Study staff will prepare both forearms to assess tick behavior on untreated and
treated forearms.
• Study staff will place ticks on the untreated and treated arms.
• Study staff will monitor and record tick movement and the start and stop times for
each exposure period.
• Study staff and study director will closely monitor the testing, and data recording.
• Alternate subjects will be enrolled to ensure adequate sample size.
• A Quality Assurance Unit will be in place to monitor all study activities and data
collection.
• There will be two calendar days in between test days for subjects participating in
more than 1 test.
(c) What QA methods are proposed?
"Arctec will be a member of the UK GLP compliance monitoring programme (GLP-
CMP) before any trial procedures are undertaken. The UK is part of the OECD GLP
Mutual Acceptance of Data agreement with the USEPA. Under this agreement, UK
laboratories that are members of the UK GLP-CMP have met the US GLP burden of
compliance, and data produced as such will be accepted as would data generated in
the US under the USEPA GLP regulations.
"The UK GLP-CMP is run by the UK GLP Monitoring Authority (UK GLPMA). The
GLPMA will inspect the test facility every 12-30 months, including an audit of
completed and on-going studies to check that the principles of GLP have been complied
Page 32 of 70
-------
with. Under the UK GLPMA, the QA unit will be a person employed by arctec, who is
independent of study conduct and directly responsibly to management. The QA
representative will conduct critical phase inspections at intervals adequate to ensure
study integrity, and maintain written and signed records of each inspection. Records
shall identify the study and include the date of the inspection, the phase inspected, the
individual conducting the inspection, positive and negative findings, actions
recommended and taken to resolve negative findings, the scheduled date for re-
inspection (if any), and the date(s) the findings are reported. All inspection findings
will be reported to management and the Study Director. Any problems, amendments or
deviations discovered shall be brought to the attention of the sponsor, Study Director
and management immediately. The QA representative will review the final reports for
accuracy and compliance with GLPs and the protocol. A signed QA statement will be
included in the final report that lists the phase inspections that were conducted, their
dates, and the dates the findings were reported to management and the Study Director.
Under UK GLPMA, the responsibility for producing a statement of compliance rests
with the Study Director. " (pg. 8, §3).
(d) How will uncertainty be addressed? Will point estimates be accompanied by
measures of uncertainty?
Sources of variation include tick species, source of ticks, tick activity, and
attractiveness of subjects. These uncertainties will be addressed by each subject
serving as their own control for qualifying ticks, using a single source for a single tick
species, qualifying ticks to ensure they are questing, and using the lowest mCPT for
the three tick species for the duration of efficacy.
3. Subject Selection
3.1 Representativeness of Sample
The population of repellent users is presumed to be diverse in age, gender, physical size,
general health, attractiveness to biting insects, and other characteristics. EPA
recommended that the protocol state explicitly that eligible subjects will be selected to be
representative by gender.
(a) What is the population of concern?
The population of concern is people who would purchase and use skin-applied insect
repellents.
(b) From what populations will subjects be recruited?
Volunteers will be recruited from the London area, specifically from the area
including and surrounding the London School of Hygiene and Tropical Medicine.
Page 33 of 70
-------
EPA recommends that the protocol be amended to expand the recruitment area (i.e.,
include the London area), add a phone number that people can use to express an
interest in participating, and describe in detail the recruitment process. For example,
where will advertisements be posted and for how long? What steps will arctec take to
ensure that the recruited population is representative of the general public/users of
skin-applied insect repellents? Who will make contact to with people who express an
interest in participating?
Participants will be recruited from a pool of interested candidates who meet the
inclusion and exclusion criteria in Section 5.2 of the protocol, as outlined below (EPA
recommendations incorporated):
Inclusion criteria
• Able to understand and comply with the study procedures
• Able to speak and understand English
• Able and willing to give fully informed consent
• Aged 18 to 65 years
• Consider themselves to be in good general health
o no serious cardiac disorder (whether active or inactive)
o no serious respiratory disorder
o no compromised immune system
o no history of anaphylaxis
• Non-smokers or willing to refrain from smoking for 24 hours prior to and
during each test
Exclusion criteria
• Localised skin disorders affecting the forearms where the product will be
applied (including but not restricted to open wounds, eczema, psoriasis,
dermatitis or open wounds)
• Participation in another clinical intervention study (excluding biting insect
challenge studies*) in the previous 3 months
• Participation in another biting insect challenge study* in the previous 72
hours
• Known or suspected history of tick bite paralysis or tick bite allergies
• Known phobia of ticks or tick bites
• Known history of anaphylaxis as a result of insect bite
• Known allergy or sensitivity to Oil of Lemon Eucalyptus or any other
repellent ingredients
• Employees and spouses of employees of the Study Sponsor (Citrefine, Inc.),
arctec, and those directly line managed by Professor James Logan (study
director)
• Students of the study director, principal investigator, or any other faculty
members involved in this study
(c) Are expected participants representative of the population of concern? If not,
why not?
Page 34 of 70
-------
The participants should be representative of the population of concern, with EPA's
recommendations about recruitment addressed. Further, EPA recommends that arctec
strive to balance the number of males and females consented to participate in the
study.
(d) Can the findings from the proposed study be generalized beyond the study
sample?
Yes.
3.2 Equitable Selection of Subjects
(a) What are the inclusion/exclusion criteria? Are they complete and appropriate?
The inclusion/exclusion criteria are complete and appropriate assuming EPA's
comments, identified in red below, are incorporated. (Section 5.2).
Inclusion criteria
• Able to understand and comply with the study procedures
• Able to speak and understand English
• Able and willing to give fully informed consent
• Aged 18 to 65 years
• Consider themselves to be in good general health
o no serious cardiac disorder (whether active or inactive)
o no serious respiratory disorder
o no compromised immune system
o no history of anaphylaxis
• Non-smokers or willing to refrain from smoking for 24 hours prior to and
during each test
Exclusion criteria
• Localised skin disorders affecting the forearms where the product will be
applied (including but not restricted to open wounds, eczema, psoriasis,
dermatitis or open wounds)
• Participation in another clinical intervention study (excluding biting insect
challenge studies*) in the previous 3 months
• Participation in another biting insect challenge study* in the previous 72
hours
• Known or suspected history of tick bite paralysis or tick bite allergies
• Known phobia of ticks or tick bites
• Known history of anaphylaxis as a result of insect bite
• Known allergy or sensitivity to Oil of Lemon Eucalyptus or any other
repellent ingredients
Page 35 of 70
-------
• Employees and spouses of employees of the Study Sponsor (Citrefine, Inc.),
arctec, and those directly line managed by Professor James Logan (study
director)
• Students of the study director, principal investigator, or any other faculty
members involved in this study
(b) What, if any, is the relationship between the investigator and the subjects?
None. EPA has suggested that the protocol exclude from participation employees and
spouses of employees of the Study Sponsor (Citrefine, Inc.), arctec, and those directly
line managed by Professor James Logan (study director). In addition, EPA
recommends excluding students of the study director, principal investigator, or any
other faculty members involved in this study.
(c) Are any potential subjects from a vulnerable population?
The protocol does not propose to target recruitment to vulnerable populations.
(d) What process is proposed for recruiting and informing potential subjects?
"Volunteers will be recruited through emails, posters, and social media to staff and
students ofLSHTM. All adverts, whether sent by email or on posters will contain a
brief description of the trial methods, an invitation to take part and an email address
with which to contact the Trial Coordination Centre (Appendix 14.3).
" Current repellent product labels are in English and the language that someone
speaks does not directly affect attractiveness to ticks. To target users familiar with
and that understand the product labels, we will be recruiting English speaking
participants. This research does not offer any benefit to the participants, so limiting
recruitment to English speakers will not result in equity-of-access issues.
" Volunteers will be fully informed before the study and it will be made clear that they
can withdraw fi'om the study at any time. Volunteers will be given and asked to read
the Participant Information Sheet and Product Information Sheet which describes the
tests which they will take part in, and a consent form which must be signed before the
test begins.
"Volunteers will be consented prior to consumer dose determination being
undertaken or being exposed to ticks. Volunteers will be given the option to consent
to the determination of the consumer dose and or to the determination of the CPT of
one or more insect tick species.
"Volunteers who do not meet the criteria for eligibility will be excluded. Individuals
who express an interest in participating in response to the recruitment materials will
be contacted by telephone or e-mail to determine whether they meet the basic
inclusion criteria. If they are interested in enrolling in the study, they will be given a
time, date and location to meet with arctec stafffor a private face-to-face meeting to
learn more about the study and their potential role in it, go over the
Page 36 of 70
-------
inclusion/exclusion criteria and receive answers to any questions they may have.
Volunteers who do not meet the criteria for eligibility will be excluded. The volunteer
will complete an eligibility questionnaire and discuss their answers with the
researchers if necessary. Contact information is included on the consent form for any
individual who has additional questions or if further clarification is desired. "
(Section 5.2)
EPA recommends that the protocol be amended to expand the recruitment area (i.e.,
include the London area), add a phone number that people can use to express an
interest in participating, and describe in detail the recruitment process. For example,
where will advertisements be posted and for how long? What steps will arctec take to
ensure that the recruited population is representative of the general public/users of
skin-applied insect repellents? Who will make contact to with people who express an
interest in participating?
EPA also recommends that the protocol be revised to include a more detailed section
covering the information to presented to participants and how it will be presented. For
example, "This meeting will cover a brief outline of the study including its purpose,
the subjects' potential role in the study, the potential length of the study on any given
test day, the identity and function of the pesticide to which they will be exposed, the
potential hazards associated with the study and steps being taken to mitigate each
hazard as addressed in the protocol, and the inclusion/exclusion criteria. The
procedures involved with the dosimetry test, as well as a 15-minute exposure interval
for the tick repellent testing will be explained and demonstrated step-bv-step to all
subjects who participate in the training. The subjects will be shown how the test
substances will be applied to their arm for the future testing."
(e) If any subjects are potentially subject to coercion or undue influence, what
specific safeguards are proposed to protect their rights and welfare?
Subjects will be recruited through emails, posters, and social media postings. There
will be no connection or communication between the researchers and the potential
subjects' employers, which minimizes the potential for coercion or undue influence.
In addition, students or employees of the study director or other faculty and
researchers involved in the study are excluded from participation. Finally, any
employees and spouses of employees of the study sponsor are excluded from
participation.
3.3 Remuneration of Subjects
(a) What remuneration, if any, is proposed for the subjects?
The protocol initially proposed " Volunteers will be paid to compensate for their time
when participating in testing. Volunteers will not be paid for consent and screening
appointments. Volunteers will be paid £20 each for taking part in the consumer dose
test, which is estimated to take approximately 30 minutes. Volunteers will be paid £60
for taking part in a repellent efficacy test which involves approximately 20 minutes
Page 37 of 70
-------
each half hour up to 10 hours." (Section 11.6) EPA recommends compensating
subjects for the consent meeting at a flat rate (e.g., £20 each for taking part in the
screening and consent meeting, which could take up to an hour). EPA also
recommends compensating subjects on an hourly basis, at close to the minimum wage
(note: in the UK, minimum wage varies by age), £7.50/hour for the lab-based
repellent testing, rounded up to the next hour.
(b) Is proposed remuneration so high as to be an undue inducement?
No.
(c) Is proposed remuneration so low that it will only be attractive to economically
disadvantaged subjects?
With EPA's recommendations addressed, no.
(d) How and when would subjects be paid?
"Participants will be paid in cash after each test session. " (Section 11.6)
4. Risks to Subjects
4.1 Risk characterization
(a) Have all appropriate prerequisite studies been performed? What do they show
about the hazards of the test material?
Subjects will be exposed to a tick repellent product (either EPA Reg. No. 84878-2 or
305-62) that is registered by the US EPA. The application of these product to the skin
of subjects in this study will be consistent with the directions for use on these
products, and therefore is determined to be no toxicological concern. The active and
inert ingredients in a currently registered formulation have undergone EPA review
and fulfilled the requirements needed for EPA registration as repellent products to be
applied to human skin.
The active component in Oil of Lemon Eucalyptus (OLE) is 65 % p-Methane-3,8-diol
(PMD). A full risk assessment for PMD technical which is used to formulate end-use
products was conducted in 1999. A 90-Day subchronic dermal study in rats (MRID
444387-10) tested PMD (98.3% pure) at increasing doses: 0, 1,000 and 3,000
mg/kg/day. The NOAEL = 1,000 mg/kg/day, and the LOAEL = 3,000 mg/kg/day.
The endpoints for NOAEL and LOAEL are based on treated skin observation,
erythema, edema, eschar, and histological observations in treated skin, increased
acanthosis and inflammation at the highest dose of 3,000 mg/kg/day. No dermal
absorption data are required for Tier I toxicity data for biochemicals. Without these
data, dermal absorption is assumed to be 100%. Risk characterization for infants and
children is based on one developmental study assessing dermal application in rabbits.
In rabbits, the NOAEL = 3,000 mg/kg/day. No LOAEL was established, and a 10-
Page 38 of 70
-------
fold safety factor is applied for risk characterization. MOAEs were not calculated
because there are no endpoints of concern for the dermal route of exposure. The
Agency concluded that there is reasonable certainty of no harm to humans from the
use of PMD in insect repellent products applied to human skin.
The toxicity profile for EPA Reg. Nos. 84878-2 and 305-62 is: toxicity category III
for acute oral, acute dermal, and acute eye irritation, and acute toxicity category IV
for acute inhalation and acute dermal irritation. Acute oral toxicity (LD50 > 2,000
mg/kg), acute dermal toxicity (LD50 > 2,000 mg/kg). For eye irritation, the product is
moderate irritant. For dermal irritation, the product is not a skin irritant, and for skin
sensitization, the product is not a skin sensitizer.
(b) What is the nature of the risks to subjects of the proposed research?
Risks to subjects include the risk of exposure to ticks, the risk of exposure to the test
material, risks related to receiving an unexpected result on a pregnancy test, and the
risk of a loss of confidentiality.
(c) How do proposed dose/exposure levels compare to the established NOAELs for
the test material?
The test material is an EPA-registered product to be used as skin applied repellent and
it will be used consistent with the Directions for Use on the product label. Because
there is no endpoint of toxicological concern for the dermal route of exposure (section
4.1 (a) above) the dose and exposure levels are lower than any NOAEL or LOAEL
for OLE, EPA considers the exposure of the subjects to the tested levels of the test
substance not to pose an unreasonable risk of adverse effects.
(d) What is the probability of each risk associated with the research? How was this
probability measured?
No numerical probability is estimated, but risks have a low probability of occurrence.
Practical steps to minimize subject risks have been described in the protocol; risks are
minimized by excluding candidates known to be hypersensitive to or phobic of tick
bites; using pathogen-free colony-raised ticks; excluding candidates known to be
sensitive to insect repellents; excluding subjects with open cuts, scrapes, skin disease
and skin problems; including medical monitoring procedures; and incorporating
procedures to keep the subjects' identities and results of pregnancy testing private,
and to permit discrete withdrawal.
4.2 Risk minimization
(a) What specific steps are proposed to minimize risks to subjects?
Participants who do not meet the eligibility criteria will be excluded from
participating. Pathogen-free ticks will be used in the study, discarded after use with a
single subject, removed immediately upon the appearance of trying to attach, and
counted in and out of the lab. Study team members will follow up with all subjects
Page 39 of 70
-------
48 hours after each instance of their participation in the study. Subjects in the
dosimetry phase will use eye protection when applying the repellent to their arms to
avoid eye irritation.
(b) What stopping rules are proposed in the protocol?
" Volunteers will be monitored throughout the duration of the tests by investigational
stafffor any adverse events. If any adverse events related to tick bites or the
repellent product are apparent at any time during the trial, testing will stop
immediately and details of how to access treatment will be offered." (Section 10.2)
"The Study Director or other designated study staff may end a particular
participant'sparticipation on a test day, at any time, for any reason(Section 11.6)
(c) How does the protocol provide for medical management of potential illness or
injury to subjects?
Study staff will follow up with subjects 48 hours after their participation to assess
whether any adverse effects have occurred.
(d) How does the protocol provide for safety monitoring?
" Volunteers will be monitored throughout the duration of the tests by investigational
stafffor any adverse events. If any adverse events related to tick bites or the
repellent product are apparent at any time during the trial, testing will stop
immediately and details of how to access treatment will be offered.'" (Section 10.2)
In addition, the study team includes a Medical Monitor.
(e) How does the protocol provide for post-exposure monitoring or follow-up? Is it
of long enough duration to discover adverse events which might occur?
Study staff will follow up with subjects 48 hours post-exposure to the test substance.
In addition, the consent form provides contact information for the Study Director
and another study team member that a subject can use in the event they suspect they
are experiencing a study-related adverse effect.
(f) How and by whom will medical care for research-related injuries to subjects be
paid for?
"The London School of Hygiene & Tropical Medicine (LSHTM) and Citrefine
International Ltd hold insurance policies which apply to this study. If you experience
harm or injury as a result of taking part in this study, you may be eligible to claim
compensation without having to prove that the LSHTM or Citrefine International Ltd
is at fault. This does not affect your legal rights to seek compensation. " (Consent
Form)
5. Benefits
Page 40 of 70
-------
(a) What benefits of the proposed research, if any, would accrue to individual
subjects?
There are no direct benefits to subjects.
(b) What benefits to society are anticipated from the information likely to be
gained through the research?
This study is designed to determine median CPT of a skin-applied tick repellent
containing OLE. The data collected in the study will be used to support product
registration. The research has societal value because people are at risk of contracting
tick-borne diseases.
(c) How would societal benefits be distributed? Who would benefit from the
proposed research?
One beneficiary will likely be the sponsor who is seeking EPA-regi strati on for skin-
applied repellent products containing OLE. Indirect beneficiaries would include the
general public who may benefit from the availability of another effective skin-
applied tick repellent.
(d) What is the likelihood that each identified societal benefits would be realized?
EPA cannot predict the outcome of the testing results; the testing could demonstrate
that the formulation is effective at providing the target level of tick repellency.
6. Risk/Benefit Balance
(a) How do the risks to subjects weigh against the anticipated benefits of the
research, to subjects or to society?
The risk mitigation measures proposed in the protocol, with EPA's
recommendations incorporated, reduce risks to subjects without reducing the
robustness of the scientific design. No reasonable opportunities to further reduce
subject risk have been overlooked. The resulting residual risk to subjects is very low.
The potential benefits from availability of a wider variety of effective skin-applied
tick repellents are likely to be realized, and the residual risks to subjects in this
proposed research are reasonable in light of the expected benefits.
7. Independent Ethics Review
(a) What IRB reviewed the proposed research?
Western Institutional Review Board
(b) Is this IRB independent of the investigators and sponsors of the research?
Page 41 of 70
-------
Yes
(c) Is this IRB registered with OHRP?
Yes
(d) Is this IRB accredited? If so, by whom?
WIRB has full AAHRPP accreditation.
(e) Does this IRB hold a Federal-Wide Assurance from OHRP?
Yes.
(f) Are complete records of the IRB review as required by 40 CFR 26.1125
provided?
Yes.
(g) What standard(s) of ethical conduct would govern the work?
This is a protocol for third-party research involving what EPA has interpreted to be
intentional exposure of human subjects to a pesticide. The study is being conducted
with the intention of submitting the resulting data to EPA under the Federal
Insecticide Fungicide and Rodenticide Act (FIFRA). Thus, the primary ethical
standards applicable to this proposal are 40 CFR 26, Subparts K and L. In addition,
the requirements of FIFRA §12(a)(2)(P) for fully informed, fully voluntary consent
of subjects apply.
8. Informed Consent
(a) Will informed consent be obtained from each prospective subject?
Yes.
(b) Will informed consent be appropriately documented, consistent with the
requirements of 40 CFR 26.1117?
Yes.
(c) Do the informed consent materials meet the requirements of 40 CFR 26.1116,
including adequate characterization of the risks and discomforts to subjects
from participation in the research, the potential benefits to the subject or
others, and the right to withdraw from the research?
With EPA's recommendations addressed, yes.
Page 42 of 70
-------
(d) What is the literacy rate in English or other languages among the intended
research subjects?
Ability to speak and read English is a requirement for participation.
(e) What measures are proposed to overcome language differences, if any, between
investigators and subjects?
N/A
(f) What measures are proposed to ensure subject comprehension of risks and
discomforts?
With EPA's recommendations addressed, there will be adequate comprehension of
risks and discomforts. The written consent form and consent meeting will address
risks and discomforts. In addition, there will be frequent opportunities to ask
questions during the consent process, as well as before and during the study. Before
a participant signs the consent form, the person conducting the meeting will ask a
series of questions based on the materials covered to ensure the participant's
comprehension.
(g) What specific procedure will be followed to inform prospective subjects and to
seek and obtain their consent?
See section 3.2(d).
(h) What measures are proposed to ensure fully voluntary participation and to
avoid coercion or undue influence?
See section 3.2(e).
9. Respect for Subjects
(a) How will information about prospective and enrolled subjects be managed to
ensure their privacy?
The subjects' identities will be protected as follows: each subject will be assigned a
code number, and only subjects' code numbers will appear on data sheets. The
subjects' names will not appear anywhere on the data sheet, or in the reports. Care
will be taken to only show participants' hands and forearms in videos taken during
the study, and particular care will be taken to not show faces or other identifiable
marks. The identity of the participants will not be disclosed with the videos.
Provision is made for discrete handling of the pregnancy testing that is required of
female subjects on the day of testing. The test results will not be disclosed to anyone
other than the test subject, the verifying female employee, and/or the Study Director.
Page 43 of 70
-------
(b) How will subjects be informed of their freedom to withdraw from the research
at any time without penalty?
Subjects will be informed about this during the informed consent meeting and at the
beginning of each test day. In addition, the consent form includes language
reminding subjects of their freedom to withdraw without penalty, and each
consented subject will receive a signed copy of this form.
(c) How will subjects who decline to participate or who withdraw from the
research be dealt with?
Subjects will be compensated if they withdraw from the research. The Study
Director or other designated study staff may end a particular participant's
participation on a test day, at any time, for any reason. If the Study Director or other
designated study staff ask a participant to withdraw from the test and they have
complied with all of their requests, or if a test participant needs to withdraw early
because of a health or emergency reason, full payment will still be made even if the
test participant has participated for less than ten hours. This will not affect payment
for any previous test days that had been completed.
If a test participant is asked to withdraw from the test because they have not
followed all their directions or if they choose to withdraw from testing early on a test
day for a non-health related or non-emergency reason, full payment will not be made
if the test participant participates in less than ten hours. Instead, they will be paid for
the number of hours worked (rounded to the nearest hour) at a rate of £7.50 per hour.
This will not affect payment for any previous test days that had been completed.
Page 44 of 70
-------
Attachment 2: Protocol Completeness Checklists
A single group trial to determine the complete protection time of an insect
repellent formulation containing 30% Citriodiol® (Oil of Lemon Eucalyptus)
against 3 species of ticks
Note: The responses in the four checklists below are based on the protocol and consent
materials that reflect EPA's recommendations.
40 CFR§ 26.1111
Criteria for Institutional Review Board (IRB) approval of research
Criterion
Y/N
Comment/Page Reference
(a)(1 )(i) Risks to subjects are minimized by using procedures which are consistent with
sound research design and which do not unnecessarily expose subjects to risk.
Y
(a)(1)(H) Risks to subjects are minimized, whenever appropriate, by using procedures
already being performed on the subjects for diagnostic or treatment purposes.
N/A
(a)(2) Risks to subjects are reasonable in relation to anticipated benefits, if any, to
subjects, and the importance of the knowledge that may reasonably be expected to
result. In evaluating risks and benefits, the IRB should consider only those risks and
benefits that may result from the research (as distinguished from risks and benefits
subjects would receive even if not participating in the research). The IRB should not
consider possible long-range effects of applying knowledge gained in the research (for
example, the possible effects of the research on public policy) as among those
research risks that fall within the purview of its responsibility.
Y
(a)(3) Selection of subjects is equitable, taking into account the purposes of the
research and the setting in which it will be conducted, and being particularly cognizant
of the special problems of research involving vulnerable populations, such as
prisoners, mentally disabled persons, or economically or educationally disadvantaged
persons.
Y
(a)(4) Informed consent will be sought from each prospective subject, in accordance
with, and to the extent required by §26.1116.
Y
(a)(5) Informed consent will be appropriately documented, in accordance with, and to
the extent required by §26.1117.
Y
(a)(6) When appropriate, the research plan makes adequate provision for monitoring
the data collected to ensure the safety of subjects.
Y
(a)(7) When appropriate, there are adequate provisions to protect the privacy of
subjects and to maintain the confidentiality of data.
Y
(b) When some or all of the subjects are likely to be vulnerable to coercion or undue
influence, additional safeguards have been included in the study to protect the rights
and welfare of these subjects.
N/A
EPA suggests that students, employees
and their spouses of PI, lab, or study
sponsor be excluded
Page 45 of 70
-------
40 CFR §26.1116
General requirements for informed consent of human subjects
Criterion
Y/N
Comment/Page Reference
No investigator may involve a human being as a subject in research covered by this
subpart unless the investigator has obtained the legally effective informed consent of
the subject or the subject's legally authorized representative
Y
An investigator shall seek such consent only under circumstances that provide the
prospective subject or the representative sufficient opportunity to consider whether or
not to participate and that minimize the possibility of coercion or undue influence
Y
The information that is given to the subject or the representative shall be in language
understandable to the subject or the representative
Y
No informed consent, whether oral or written, may include any exculpatory language
through which the subject or the representative is made to waive or appear to waive any
of the subject's legal rights, or releases or appears to release the investigator, the
sponsor, the institution or its agents from liability for negligence
Y
c
o
CO
o
(1) A statement that the study involves research, an explanation of the
purposes of the research and the expected duration of the subject's
participation, a description of the procedures to be followed, and
identification of any procedures which are experimental
Y
c
O)
c
(2) A description of any reasonably foreseeable risks or discomforts to the
subject
Y
5
o
o
(3) A description of any benefits to the subject or to others which may
reasonably be expected from the research
Y
0
_c
za
c a)
=>
8 «
""5
s s
§*
¥ -a
(4) A disclosure of appropriate alternative procedures or courses of
treatment, if any, that might be advantageous to the subject
N/A
No alternative procedures; alternative is
not to participate
(5) A statement describing the extent, if any, to which confidentiality of
records identifying the subject will be maintained
Y
(6) For research involving more than minimal risk, an explanation as to
whether any compensation and an explanation as to whether any
medical treatments are available if injury occurs and, if so, what they
consist of, or where further information may be obtained
Y
:i!
c
0 Q.
82
C —
_ CO
s-s
(7) An explanation of whom to contact for answers to pertinent questions
about the research and research subjects' rights, and whom to contact in
the event of a research-related injury to the subject
Y
(8) A statement that participation is voluntary, refusal to participate will
involve no penalty or loss of benefits to which the subject is otherwise
entitled, and the subject may discontinue participation at anytime
without penalty or loss of benefits to which the subject is otherwise
entitled
Y
0 =
o 05
(1) A statement that the particular treatment or procedure may involve
risks to the subject (or to the embryo or fetus, if the subject may become
pregnant) which are currently unforeseeable
Y
£ o "G
o =
E E ~
® L. O)
(2) Anticipated circumstances under which the subject's participation
may be terminated by the investigator without regard to the subject's
consent
Y
B o ®
(3) Any additional costs to the subject that may result from participation
in the research
N/A
No anticipated costs to the subject
« (0 o
l_ Jgi
Q. C -o
P 0
m ® s
(4) The consequences of a subject's decision to withdraw from the
research and procedures for orderly termination of participation by the
subject
Y
EPA recommends adding additional
information to protocol and consent
form about how withdrawal during field
trial will occur
j= .E ®
1
> o o
(0
(5) A statement that significant new findings developed during the course
of the research which may relate to the subject's willingness to continue
participation will be provided to the subject
Y
(6) The approximate number of subjects involved in the study
Y
(e) If the research involves intentional exposure of subjects to a pesticide, the subjects
of the research must be informed of the identity of the pesticide and the nature of its
pesticidal function.
Y
Page 46 of 70
-------
40 CFR §26.1117
Documentation of informed consent
Criterion
Y/N
Comment/Page Reference
(a) Informed consent shall be documented by the use of a written consent form
approved by the IRB and signed by the subject or the subject's legally authorized
representative. A copy shall be given to the person signing the form.
Y
(b)(1) The consent form may be a written consent document that embodies the
elements of informed consent required by 40 CFR §26.1116. This form may be
read to the subject or the subject's legally authorized representative, but in any
event, the investigator shall give either the subject or the representative adequate
opportunity to read it before it is signed; or
Y
(b)(2) The consent form may be a short form written consent document stating that
the elements of informed consent required by 40 CFR §26.1116 have been
presented orally to the subject or the subject's legally authorized representative.
When this method is used, there shall be a witness to the oral presentation. Also, the
IRB shall approve a written summary of what is to be said to the subject or the
representative. Only the short form itself is to be signed by the subject or the
representative. However, the witness shall sign both the short form and a copy of the
summary, and the person actually obtaining consent shall sign a copy of the
summary. A copy of the summary shall be given to the subject or the representative,
in addition to a copy of the short form.
Page 47 of 70
-------
40 CFR §26.1125
Submission of proposed human research for EPA review
Any person or institution who intends to conduct or sponsor human research covered by §26.1101(a)
shall, after receiving approval from all appropriate IRBs, submit to EPA prior to initiating such
research all information relevant to the proposed research specified by 40 CFR §26.1115(a). and
the following additional information, to the extent not already included:
Requirement
Y/N
Comments/Page Refs
The following Information, to the
Extent not already included:
§1125(a)
a discussion of:
(1) The potential risks to human subjects
Y
(2) The measures proposed to minimize risks to the human
subjects;
Y
(3) The nature and magnitude of all expected benefits of such
research, and to whom they would accrue
Y
(4) Alternative means of obtaining information comparable to what
would be collected through the proposed research; and
Y
(5) The balance of risks and benefits of the proposed research.
Y
§1125(b): All information for subjects and written informed consent
agreements as originally provided to the IRB, and as approved by the IRB.
Y
§1125(c): Information about how subjects will be recruited, including any
advertisements proposed to be used.
Y
Arctec must provide EPA with all
recruitment materials, including scripts
and email templates prior to initiating
the study
§1125(d): A description of the circumstances and methods proposed for
presenting information to potential human subjects for the purpose of
obtaining their informed consent.
Y
§1125(e): All correspondence between the IRB and the investigators or
sponsors.
Y
Separate file "arctec OLE IRB
correspondence and approval"
§1125(f): Official notification to the sponsor or investigator. . . that research
involving human subjects has been reviewed and approved by an IRB.
Y
Separate file "arctec OLE IRB
correspondence and approval"
all information relevant to the proposed research
specified by § 26.1115(a)
(1) Copies of
all research proposals reviewed by the IRB,
scientific evaluations, if any, that accompanied the
proposals reviewed by the IRB,
approved sample consent documents,
progress reports submitted by investigators, and reports
of injuries to subjects.
Y
(2) Minutes of IRB meetings ... in sufficient detail to show
attendance at the meetings;
actions taken by the IRB;
the vote on these actions including the number of members voting
for, against, and abstaining;
the basis for requiring changes in or disapproving research;
a written summary of the discussion of controverted issues and
their resolution.
Y
Separate file "IRB minutes"
(3) Records of continuing review activities.
N/A
(4) Copies of all correspondence between the IRB and the investigators.
Y
Separate file "arctec OLE IRB
correspondence and approval"
Page 48 of 70
-------
(5) A list of IRB members identified by name; earned degrees;
representative capacity; indications of experience such as board
certifications, licenses, etc., sufficient to describe each member's chief
anticipated contributions to IRB deliberations;
any employment or other relationship between each member and the
institution, for example, full-time employee, a member of governing panel or
board, stockholder, paid or unpaid consultant.
Y
Separate file "arctec OLE IRB
correspondence and approval"
(6) Written procedures for the IRB in the same detail as described in
§26.1108(a) and §26.1108(b).
Y
Provided to EPA previously
(7) Statements of significant new findings provided to subjects, as required
by §26.1116(b)(5).
N/A
Page 49 of 70
-------
Attachment 3: EPA's Power vs. Sample Size Calculation for Tick Repellency Studies
Objective
To determine the sample size of N subjects such that tick repellency studies have sufficient power to obtain a given
degree of precision in the estimate of median Complete Protection Time (mCPT). This precision - designated as
"K" -- will be expressed as the ratio: 95% LCLmCpT/estimated mCPT.
The simulation used to estimate varying sample sizes will require that that 95% LCLmCpT/estimated mCPT 0,
x < 0
x > 0,
x < 0
Page 50 of 70
-------
mCPT= estimated median complete protection time
95% LCLmcPT = 95% lower confidence limit on the estimated mCPT
Similarly, the degree of variation of the CPT distribution in the population will be defined as the P5MR which we
define here as the ratio between the mCPT of the 5th percentile of the population to the mCPT of the population:
P5MR = CPT 5th %iie /mCPT
where:
mCPT= median complete protection time
CPTsth %iie = 5th percentile of the distribution of CPT
Re-parameterization of Standard Weibull Equation
While the above mCPT and P5MR parameterizations of the Weibull distribution are intuitively appealing for
judging and evaluating repellent efficacy, they are non-standard parameterizations and it is necessary ~ for
comparison and simulation purposes ~ to convert these to the more standard k (shape) and 1 (scale) values. To do
this, EPA developed an equation such that interconversion between the standard (k (shape) and 1 (scale))
parameterization of the Weibull to this alternate version (with the Weibull distribution instead expressed in terms of
P5MR and mCPT). Briefly, the cumulative probability function of CPT is assumed to be a 2- parameter Weibull
distribution:
P(CPT, k, 1) = l-e~(CPT/X)K
Given that a value of the mCPT represents the median or 50th percentile of the CPT and the value of P5MR
represents the ratio of the 5%-tile of the CPT distribution to the mCPT, we can develop the following two equations
to represent the cumulative distribution functions at the median CPT and the 5th percentile CPT:
(mCPT\K
P(mCPT,K,X)= 1 — e ^ ^ > =0.5 (median)
!P5MRXmCPT\K
P(P5MR x mCPT,K,A) = 1 — e ^ ^ > = 0.05 (5th percentile)
Algebraically solving the equations above (see Appendix 14.2A for full derivation), we develop expressions for k
and 1:
ln(0.95)
k = In
ln(0.5)
1 . r mCPTK
- X In -,
/ \n(P5MR)
X = e~K I ln(0.5)
Table A1 below compares these two parameterizations for the example PDF and CDF distributions shown in
Figures A1 and A2, respectively, for the k and 1 parameterizations shown there, illustrating the conversion to this
new parameterization:
Table A1. Re-parameterization of Weibull Distribution Parameters from Traditional (k, X) to Revised (P5MR,
mCPT) for Example Weibull Distributions Appearing in Figures A1 and A2.
Parameterization Scheme
Description/Comments
Traditional
Revised
Scale (X)a
Shape (k)
mCPTb
P5MRcd
1
0.5
0.480453
0.005476
- k values of less than 1 indicate a crossing rate
decreases over time, and defective items fail early or
are otherwise removed from the population.
1
1
0.693147
0.074001
- k values equal to 1 indicate a constant crossing rate
over time possibly suggesting crossing is due to
random external events.
- Here, the Weibull distribution reduces to the
"exponential" distribution;
- Note that mCPT here = 0.693 = ln(2)
1
1.5
0.78322
0.176261
- k values greater than 1 suggests that the crossing rate
increases over time, as when there is an "aging"
process or components are more likely to fail over
time.
1
5
0.92932
0.594083
Page 51 of 70
-------
aThe Weibull scale parameter is the 63.2 percentile of the distribution. If the scale parameter is 1, then this
means that 63.2% of the observed values will be smaller than 1. Note in the CDF in Figure A2, as a consequence,
that all A=1 distributions intersect at the 63.2 percentile.
b mCPT = [ln(2)*exp(K *ln(A))](1/K)
c P5MR = exp(ln(ln(0.95)/ln(0.5))/K)
d Note that as k increases, the P5MR value becomes larger, indicating that the values at the 5th percentile
approach the values present at the 50th percentile, and the PDF becomes tighter and more peaked. K values of
between 3 and 4 often lead to distributions that appear normal.
Page 52 of 70
-------
Figure Al. Probability Density Function (PDF) for Weibull Plot with 1 (scale) =1 and k (shape) ranging from 0.5 to
5
00 6.6 10 t6 20 2*6
Figure A2. Cumulative Distribution Function (CDF) for Above Weibull PDF
Weibull(2-Parameter) Cumulative distribution function
o'o o's to t's 10 25
Page 53 of 70
-------
Figure A3: Histograms of CPT distributions for various CPTs and P5MRs (assume CPTs are Weibull distributions)
weibull distribution: median=2
P5MR= 0.2
TlTTTTTTTlTlTTTTTT-rn^
P5MR= 0.4
P5MR= 0.5
TlTTTh-r>^
TTTTt-h^,
CPT
weibull distribution: median=4
6-
4-
2-
4-
2-
8-
6-
4-
2-
6-
4-
2-
P5MR= 0.2
TTITTTTTTTTTTTTTTTTttt^
P5MR= 0.4
P5MR= 0.5
Trrrrrrm^
-^rrfTf
5
CPT
~~i r
9 10
weibull distribution: median=6
weibull distribution: median=8
P5MR= 0.2
TTTTTTTTTITTlTlTrTri
P5MR= 0.3
^rfftrTllT
P5MR= 0.5
T11TTTTtt-.-^_
P5MR= 0.2
TTTTTTTTTTTlTTTTTln~rm
P5MR= 0.3
^^ttttTTIT
P5MR= 0.5
TITTTTTt'-'-—
Th"h-^-
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
CPT
_l 1 f—( ! ! ! ! 1 1 r
0 1 2 3 4 5 6 7 8 9 10111213141516171819 20
CPT
-------
An example of the (varied) kinds of distributional "shapes" associated with various parameterizations is shown in
Figure A3 as histograms of the CPT. More specifically, Figure A3 presents the CPT distributions with different
medians and values of P5MR (ratio 5%-tile/mCPT). These present the CPT distributions with different mCPTs (2-,
4-, 6-, and 8- hrs) and values of the P5MR ratio (P5MR= 0.2, 0.3, 0.4, 0.5, and 0.6) for the (assumed) Weibull
Distributions. As seen in Figure A3, larger mCPTs are associated with a shift in the distribution toward the right. In
addition ~ and importantly ~ smaller P5MR values in this range are associated with "flatter" distributions and larger
P5MRs are associated with more "peaked" distributions, with these more peaked distributions showing a greater
percentage of the distribution centered around the median. From a regulatory perspective, a CPT distribution with a
larger P5MR is more desirable than a CPT distribution with smaller P5MR since this means that a greater
percentage of the user population experiences an actual CPT closer to the (advertised) mCPT. Further, it could be
argued from a public policy perspective that a large variability in CPT in the population for a given repellent is not a
desirable characteristic, and does not accurately portray or indicate any "expected" mCPT on the part of the
consumer.
OPP staff have judged what might be considered reasonable values for input parameters (precision of the estimated
mCPT and variability in CPT in (or among) users of the tested product) in order to estimate required number of test
subjects to achieve a desired set of aims regarding precision around the estimate of the mCPT. These judgments are
based in part on the data of a study4 and in part on general thoughts regarding consumer and other expectations with
respect to product efficacy. Specifically, EPA has estimated the power associated with various sample sizes where
power -- as defined here - is the probability that the ratio of the (95% LCLmCpT)/(estimated mCPT) is equal or
greater than a given acceptable K (a scalar which measures the precision of the estimates in estimating the mCPT).
Such tick repellency study design power depends on:
• Number of test subjects
o The larger the number of test subjects, the greater the power
• (The required) precision (K) for estimated mCPT
o The precision of an estimated mCPT from a study is expressed by the value of the ratio 95%
LCLmcpi/estimated mCPT. The value of ratio is in the interval (0, 1).
o K is the smallest acceptable value of the ratio 95% LCLmCpT/estimated mCPT for a given trial
to be considered a "success", and conceptually represents an inverse of precision ("tightness")
in the estimate of the mCPT: a larger K represents a greater "tightness" around the estimated
mCPT. As K is chosen to be smaller, there is a greater probability that ratio 95%
LCLmcpi/estimated mCPT > K (and the trial is considered to be a "success" in the power
calculation)
• P5MR
o P5MR = ratio of the 5th percentile/mCPT
o As the variation (dispersion or spread) of the distribution of CPT in the population becomes
smaller, the 95% confidence interval of the estimated mCPT also becomes narrower (i.e. the
95% LCLmcPT is closer to the estimated mCPT and the mCPT is better estimated, certeris
paribus). Therefore, a smaller variation in the distribution of CPT will result in a larger
P5MR and a higher probability that the ratio 95% LCLmCpT/estimated mCPT > K. A CPT
distribution with greater P5MR is generally more desirable than a CPT distribution with
smaller P5MR.
Ideally, a tick repellency study will be designed to have a sufficient number of test subjects such that one can
have reasonable assurance that there is adequate power (defined here as a high probability that the ratio 95%
LCL/estimated mCPT > K) given a shape/spread of the CPT distribution in the population. This shape/spread of
the CPT in the population is defined by the P5MR.
Brief Description of the Conduct of a Ticks Repellent Study
In the tick repellency studies, each test subject has 3 lines drawn on the testing arm, with a distance of 3 cm
between any two adjacent lines as shown in the Figure 2. Product will be applied from the boundary line (at the
4 See Appendix 14.2B for Weibull parameters fit to CPT data of a tick study
Page 55 of 70
-------
wrist) to the elbow. One tick at a time will be released at the release line. A "crossing" is recorded if the test
organism crosses the boundary line at least 3 cm into the treated area within 3 minutes, and remains in the
treated area for at least one minute.
A crossing is a confirmed crossing if it is followed by another crossing within 30 minutes. For subjects who
receive confirmed crossings, the CPTs are set as 0 if the first confirmed crossing occurs during the test of 1 tick
immediately following product application; otherwise, the CPTs are rounded down to the nearest quarter hour
(i.e., the starting time of the testing interval in which the first confirmed crossing occurs). We approximated
that it would take 3 minutes to test 1 tick in each 15-minute interval. For those subjects for which there are no
confirmed crossings through the end of the testing day, CPTs are considered to be right censored at a time that
is rounded down to the nearest half hour.
Description of (Comvuter) Simulation Procedure:
To start the simulated study trials, 4000 datasets were created with each dataset consisting of 10 data points
(representing CPTs of 10 subjects) that were generated randomly from a Weibull distribution with a median
CPT=2 and ratio of the 5%-tile/median P5MR= 0.2. If the randomly generated CPTs for the 10 subjects are <
3, 4-18, 19-33, 34-48, 49-63, ... -minutes, the CPTs are set to be 0-, 0.25-, 0.5-, 0.75-, 1-hours..., respectively,
to simulate the study design in which each study participant would take about 3 minutes to test 1 tick for every
15 minutes until the first confirmed crossing. If the randomly generated CPTs are greater than 723 minutes,
they are considered in the calculation to be (right) censored at 12 hours.
After generating the CPTs as described in the previous paragraph, the Kaplan Meier Estimator is used to
estimate the mCPT and its 95% CI for each of the 4000 (10-person) datasets. The proportion of datasets in
which the ratio of 95% LCLmcpi/mCPT > K as 0.6 is considered to be the "power" of the study design. More
specifically: if the value of 95% LCL/mCPT > 0.6 is considered a "success", the power is calculated as the
proportion of successes in the 4000 datasets consisting of 10 data points each.
The process described in previous paragraph is then repeated for each combination of different mCPT = 2, 4, 6,
8, and 10 hours; P5MR = 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, and 0.8; sample size per dataset = 10, 11, 12 ... 30; and the
lowest acceptable K = 0.6, 0.7, and 0.8; all assuming that CPT follows Weibull distributions.
Results of Simulation
Tables A2, A3, ... A7 present the power estimates from simulations in which the data were randomly generated
from Weibull distributions for K = 0.6, 0.7, and 0.8. These are shown for various values of mCPT (ranging from
2 to 10 hours), P5MR (ranging from 0.2 to 0.8), and Sample Size (ranging from 10 to 30). As described earlier,
K reflects a measure the precision of the estimate of mCPT with larger K values representing tighter estimates.
For example, the K value of 0.6 requires that the 95% LCL on an estimated median protection of 10 hours be no
less than 6 hours (for a "success") while a K value of 0.8 requires that the 95% LCL on that same median
protection time be no less than 8 hours. A required precision of a K of 0.8, then, requires a more precise
estimate of the mCPT than a K of 0.6 for this trial to be considered a "success" in the power calculation.
As can be seen within each Table, the power of a study to achieve a given acceptable ratio K value (e.g., 0.6,
0.7, or 0.8 representing 95% LCLmCPT/mCPT) value increases as the assumed P5MR value of the distribution
increases (for example, from 0.2 to 0.8) or as the sample size increases (from 10 to 20 or from 21 to 30). This is
expected since a tighter (or more "peaked") distributions (as evidenced by a larger P5MR value) will require
fewer random "draws" to accurately estimate the mCPT. Across the Tables, we also see that as the acceptable
K value increases from 0.6 to 0.8, the power of a study to achieve "95% LCLmCPT/mCPT > K" decreases
since stricter requirements for a "success" are being levied.
Page 56 of 70
-------
Table A2: Power when the lowest acceptable ratio 95% LCLmCPT/mCPT = 0.6
(Weibull distribution)
median
P5MR
Sample Size
10
11
12
13
14
15
16
17
18
19
20
2
0.2
0.045
0.198
0.150
0.356
0.280
0.245
0.426
0.387
0.548
0.512
0.435
0.3
0.109
0.361
0.313
0.577
0.496
0.467
0.664
0.642
0.800
0.747
0.693
0.4
0.231
0.572
0.514
0.786
0.719
0.708
0.860
0.845
0.929
0.913
0.886
0.5
0.410
0.780
0.738
0.924
0.890
0.883
0.960
0.959
0.986
0.982
0.977
0.6
0.638
0.932
0.914
0.986
0.975
0.979
0.993
0.996
0.995
0.997
0.997
0.7
0.871
0.993
0.988
0.993
0.994
0.995
0.995
0.996
0.979
0.990
0.994
0.8
0.979
0.973
0.988
0.946
0.963
0.941
0.924
0.947
0.874
0.907
0.941
4
0.2
0.037
0.175
0.130
0.328
0.257
0.222
0.405
0.360
0.523
0.474
0.399
0.3
0.097
0.340
0.290
0.560
0.477
0.437
0.655
0.621
0.789
0.735
0.687
0.4
0.213
0.542
0.505
0.769
0.712
0.685
0.855
0.827
0.934
0.902
0.893
0.5
0.402
0.757
0.734
0.918
0.895
0.873
0.962
0.955
0.989
0.979
0.980
0.6
0.648
0.924
0.918
0.979
0.980
0.973
0.996
0.995
0.999
0.999
0.998
0.7
0.871
0.992
0.992
0.999
0.999
1.000
1.000
1.000
1.000
1.000
1.000
0.8
0.987
0.999
1.000
0.998
0.999
1.000
0.998
0.999
0.993
0.998
0.998
6
0.2
0.038
0.162
0.129
0.316
0.252
0.213
0.387
0.343
0.512
0.463
0.389
0.3
0.093
0.325
0.273
0.540
0.463
0.421
0.642
0.601
0.775
0.723
0.680
0.4
0.203
0.529
0.499
0.762
0.703
0.677
0.844
0.826
0.931
0.899
0.890
0.5
0.398
0.749
0.729
0.914
0.894
0.868
0.962
0.950
0.989
0.980
0.977
0.6
0.637
0.925
0.916
0.982
0.982
0.976
0.997
0.996
1.000
0.999
0.999
0.7
0.870
0.992
0.990
0.999
0.999
0.999
1.000
1.000
1.000
1.000
1.000
0.8
0.987
0.999
1.000
1.000
1.000
1.000
1.000
1.000
0.999
1.000
1.000
8
0.2
0.120
0.200
0.182
0.337
0.291
0.229
0.407
0.353
0.526
0.466
0.405
0.3
0.116
0.327
0.290
0.538
0.471
0.417
0.640
0.598
0.773
0.723
0.676
0.4
0.202
0.523
0.491
0.754
0.700
0.672
0.845
0.816
0.930
0.897
0.885
0.5
0.390
0.745
0.724
0.913
0.890
0.865
0.960
0.950
0.989
0.980
0.978
0.6
0.629
0.923
0.915
0.981
0.981
0.974
0.996
0.995
1.000
0.999
0.999
0.7
0.865
0.992
0.990
0.998
0.999
0.999
1.000
1.000
1.000
1.000
1.000
0.8
0.986
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
10
0.2
0.371
0.374
0.407
0.467
0.481
0.369
0.558
0.477
0.652
0.548
0.532
0.3
0.330
0.450
0.446
0.614
0.585
0.491
0.712
0.649
0.817
0.753
0.725
0.4
0.338
0.576
0.566
0.779
0.746
0.690
0.866
0.831
0.937
0.899
0.894
0.5
0.442
0.754
0.739
0.918
0.896
0.867
0.961
0.953
0.988
0.980
0.978
0.6
0.637
0.920
0.914
0.980
0.981
0.974
0.997
0.996
1.000
0.999
0.999
0.7
0.865
0.992
0.991
0.999
0.999
0.999
1.000
1.000
1.000
1.000
1.000
0.8
0.986
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
NOTE: Yellow indicates power > 0.8; orange indicates power > 0.9
Page 57 of 70
-------
Table A3: Power when the lowest acceptable ratio 95% LCLmCPT/mCPT = 0.6
(Weibull distribution)
median
P5MR
Sample Size
21
22
23
24
25
26
27
28
29
30
2
0.2
0.610
0.556
0.727
0.657
0.806
0.736
0.720
0.815
0.777
0.855
0.3
0.840
0.808
0.906
0.867
0.950
0.915
0.919
0.952
0.947
0.970
0.4
0.959
0.948
0.982
0.970
0.991
0.985
0.989
0.995
0.993
0.997
0.5
0.996
0.992
0.998
0.996
0.999
0.999
0.998
0.998
1.000
0.999
0.6
0.997
0.998
0.995
0.999
0.996
0.996
0.998
0.992
0.997
0.998
0.7
0.983
0.987
0.967
0.979
0.979
0.964
0.980
0.947
0.968
0.971
0.8
0.865
0.901
0.819
0.850
0.869
0.810
0.859
0.760
0.806
0.837
4
0.2
0.586
0.528
0.705
0.638
0.789
0.721
0.694
0.808
0.763
0.857
0.3
0.824
0.807
0.902
0.865
0.942
0.918
0.907
0.957
0.939
0.974
0.4
0.958
0.952
0.978
0.973
0.989
0.985
0.987
0.996
0.993
0.998
0.5
0.996
0.994
0.998
0.998
0.999
1.000
0.999
1.000
1.000
1.000
0.6
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.7
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.999
1.000
0.999
0.8
0.994
0.997
0.989
0.996
0.993
0.991
0.997
0.986
0.992
0.992
6
0.2
0.575
0.517
0.689
0.626
0.782
0.715
0.683
0.804
0.750
0.852
0.3
0.819
0.794
0.891
0.861
0.942
0.917
0.905
0.955
0.939
0.974
0.4
0.955
0.949
0.977
0.969
0.990
0.985
0.986
0.996
0.994
0.997
0.5
0.997
0.992
0.999
0.998
0.999
1.000
0.999
1.000
1.000
1.000
0.6
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.7
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.8
1.000
1.000
0.999
1.000
1.000
1.000
1.000
0.998
0.999
0.998
8
0.2
0.573
0.519
0.690
0.627
0.783
0.713
0.681
0.801
0.747
0.850
0.3
0.810
0.802
0.889
0.859
0.938
0.913
0.900
0.954
0.935
0.972
0.4
0.953
0.948
0.976
0.973
0.991
0.986
0.987
0.996
0.994
0.997
0.5
0.997
0.993
0.998
0.997
0.999
1.000
1.000
1.000
1.000
1.000
0.6
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.7
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.8
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
10
0.2
0.652
0.633
0.743
0.716
0.819
0.773
0.721
0.840
0.784
0.883
0.3
0.836
0.827
0.907
0.881
0.946
0.922
0.908
0.965
0.944
0.978
0.4
0.956
0.950
0.978
0.975
0.990
0.986
0.988
0.995
0.994
0.998
0.5
0.997
0.993
0.999
0.998
0.999
1.000
1.000
1.000
1.000
1.000
0.6
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.7
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.8
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
Page 58 of 70
-------
Table A4: Power when the lowest acceptable ratio 95% LCLmCPT/mCPT = 0.7
(Weibull distribution)
median
P5MR
Sample Size
10
11
12
13
14
15
16
17
18
19
20
2
0.2
0.013
0.077
0.048
0.165
0.116
0.089
0.198
0.169
0.287
0.246
0.184
0.3
0.036
0.172
0.132
0.320
0.262
0.216
0.393
0.346
0.520
0.452
0.401
0.4
0.096
0.314
0.275
0.516
0.454
0.402
0.624
0.573
0.741
0.684
0.652
0.5
0.198
0.504
0.484
0.717
0.681
0.622
0.826
0.776
0.909
0.850
0.862
0.6
0.403
0.717
0.726
0.882
0.872
0.833
0.949
0.921
0.975
0.958
0.968
0.7
0.671
0.908
0.926
0.970
0.975
0.964
0.989
0.988
0.978
0.985
0.993
0.8
0.922
0.969
0.983
0.944
0.963
0.940
0.924
0.946
0.874
0.907
0.940
4
0.2
0.009
0.061
0.038
0.134
0.090
0.065
0.167
0.129
0.240
0.206
0.149
0.3
0.027
0.139
0.105
0.269
0.213
0.174
0.339
0.293
0.456
0.404
0.343
0.4
0.071
0.271
0.229
0.469
0.403
0.353
0.581
0.520
0.708
0.654
0.609
0.5
0.169
0.466
0.432
0.709
0.646
0.604
0.796
0.761
0.899
0.856
0.838
0.6
0.357
0.708
0.689
0.888
0.865
0.833
0.951
0.932
0.983
0.972
0.972
0.7
0.635
0.922
0.913
0.980
0.983
0.972
0.996
0.995
0.999
1.000
0.999
0.8
0.910
0.995
0.996
0.998
0.998
1.000
0.998
0.999
0.993
0.998
0.998
6
0.2
0.013
0.060
0.038
0.130
0.089
0.066
0.163
0.121
0.234
0.205
0.147
0.3
0.026
0.139
0.098
0.263
0.212
0.169
0.333
0.283
0.446
0.396
0.326
0.4
0.069
0.265
0.224
0.466
0.395
0.341
0.564
0.512
0.697
0.642
0.589
0.5
0.158
0.458
0.419
0.697
0.631
0.592
0.788
0.755
0.897
0.851
0.831
0.6
0.347
0.697
0.678
0.885
0.855
0.820
0.943
0.928
0.983
0.966
0.967
0.7
0.628
0.912
0.906
0.979
0.979
0.970
0.996
0.995
1.000
0.999
0.998
0.8
0.906
0.996
0.996
0.999
0.999
1.000
1.000
1.000
0.999
1.000
1.000
8
0.2
0.098
0.104
0.101
0.155
0.131
0.086
0.193
0.140
0.249
0.201
0.161
0.3
0.052
0.141
0.116
0.267
0.212
0.173
0.334
0.284
0.443
0.392
0.329
0.4
0.072
0.261
0.221
0.463
0.388
0.341
0.559
0.510
0.698
0.642
0.587
0.5
0.161
0.457
0.420
0.689
0.626
0.587
0.784
0.754
0.896
0.848
0.827
0.6
0.350
0.699
0.674
0.888
0.858
0.823
0.945
0.929
0.982
0.968
0.964
0.7
0.625
0.914
0.906
0.980
0.979
0.972
0.996
0.996
1.000
0.999
0.999
0.8
0.905
0.995
0.996
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
10
0.2
0.352
0.291
0.346
0.315
0.358
0.248
0.387
0.289
0.436
0.329
0.327
0.3
0.277
0.284
0.308
0.371
0.373
0.272
0.453
0.367
0.544
0.449
0.419
0.4
0.225
0.338
0.335
0.502
0.469
0.379
0.613
0.539
0.723
0.658
0.617
0.5
0.226
0.475
0.453
0.705
0.644
0.595
0.795
0.761
0.902
0.850
0.839
0.6
0.360
0.696
0.685
0.890
0.862
0.828
0.947
0.930
0.982
0.969
0.965
0.7
0.626
0.921
0.913
0.978
0.980
0.973
0.995
0.995
1.000
0.999
0.998
0.8
0.911
0.997
0.996
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
Page 59 of 70
-------
Table A5: Power when the lowest acceptable ratio 95% LCLmCPT/mCPT = 0.7
(Weibull distribution)
median
P5MR
Sample Size
21
22
23
24
25
26
27
28
29
30
2
0.2
0.327
0.267
0.420
0.353
0.532
0.435
0.400
0.516
0.462
0.585
0.3
0.569
0.514
0.666
0.609
0.753
0.700
0.663
0.784
0.713
0.842
0.4
0.763
0.759
0.855
0.832
0.904
0.886
0.857
0.934
0.898
0.951
0.5
0.917
0.924
0.952
0.948
0.976
0.972
0.960
0.986
0.975
0.989
0.6
0.983
0.988
0.986
0.991
0.993
0.993
0.994
0.990
0.995
0.998
0.7
0.983
0.986
0.967
0.979
0.979
0.964
0.980
0.947
0.968
0.971
0.8
0.865
0.901
0.819
0.850
0.869
0.810
0.859
0.760
0.806
0.837
4
0.2
0.279
0.225
0.371
0.299
0.467
0.376
0.340
0.462
0.400
0.533
0.3
0.513
0.455
0.624
0.558
0.727
0.652
0.619
0.746
0.687
0.808
0.4
0.752
0.722
0.846
0.804
0.907
0.873
0.853
0.928
0.893
0.953
0.5
0.925
0.923
0.960
0.951
0.980
0.973
0.972
0.990
0.986
0.995
0.6
0.992
0.990
0.997
0.994
0.998
1.000
0.999
1.000
1.000
1.000
0.7
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.999
1.000
0.999
0.8
0.994
0.997
0.989
0.996
0.993
0.991
0.997
0.986
0.992
0.992
6
0.2
0.279
0.222
0.369
0.285
0.461
0.369
0.328
0.450
0.392
0.525
0.3
0.501
0.447
0.620
0.547
0.720
0.638
0.610
0.736
0.673
0.800
0.4
0.737
0.714
0.836
0.795
0.901
0.863
0.839
0.922
0.888
0.948
0.5
0.922
0.913
0.958
0.947
0.978
0.972
0.971
0.989
0.984
0.993
0.6
0.990
0.990
0.997
0.995
0.998
0.999
0.998
1.000
1.000
1.000
0.7
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.8
1.000
1.000
0.999
1.000
1.000
1.000
1.000
0.998
0.999
0.998
8
0.2
0.276
0.226
0.370
0.296
0.456
0.369
0.329
0.455
0.391
0.528
0.3
0.495
0.442
0.614
0.545
0.719
0.636
0.607
0.731
0.676
0.799
0.4
0.745
0.712
0.840
0.794
0.898
0.867
0.844
0.920
0.889
0.948
0.5
0.921
0.916
0.959
0.946
0.979
0.971
0.969
0.990
0.985
0.993
0.6
0.992
0.988
0.997
0.995
0.998
0.999
0.998
1.000
1.000
1.000
0.7
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.8
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
10
0.2
0.402
0.385
0.461
0.434
0.537
0.484
0.413
0.551
0.477
0.618
0.3
0.550
0.519
0.651
0.608
0.740
0.672
0.630
0.758
0.698
0.825
0.4
0.753
0.733
0.847
0.813
0.902
0.869
0.848
0.926
0.891
0.952
0.5
0.925
0.914
0.961
0.947
0.980
0.971
0.972
0.989
0.985
0.994
0.6
0.993
0.988
0.996
0.996
0.998
0.999
0.999
1.000
1.000
1.000
0.7
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.8
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
Page 60 of 70
-------
Table A6: Power when the lowest acceptable ratio 95% LCLmCPT/mCPT = 0.8
(Weibull distribution)
median
P5MR
Sample Size
10
11
12
13
14
15
16
17
18
19
20
2
0.2
0.003
0.026
0.008
0.051
0.030
0.020
0.056
0.049
0.082
0.070
0.044
0.3
0.009
0.050
0.031
0.099
0.078
0.051
0.131
0.087
0.192
0.137
0.114
0.4
0.022
0.098
0.078
0.175
0.160
0.112
0.257
0.176
0.335
0.254
0.249
0.5
0.054
0.173
0.171
0.303
0.309
0.208
0.435
0.308
0.544
0.419
0.434
0.6
0.129
0.296
0.335
0.492
0.505
0.387
0.649
0.526
0.745
0.632
0.668
0.7
0.307
0.532
0.585
0.729
0.758
0.657
0.857
0.793
0.905
0.858
0.893
0.8
0.618
0.818
0.867
0.894
0.922
0.870
0.908
0.912
0.868
0.891
0.932
4
0.2
0.002
0.019
0.006
0.039
0.023
0.014
0.039
0.032
0.069
0.050
0.030
0.3
0.005
0.042
0.023
0.088
0.058
0.039
0.102
0.081
0.162
0.131
0.090
0.4
0.014
0.090
0.060
0.182
0.136
0.103
0.231
0.190
0.328
0.286
0.212
0.5
0.047
0.190
0.148
0.353
0.284
0.248
0.435
0.393
0.558
0.519
0.443
0.6
0.114
0.381
0.323
0.606
0.521
0.496
0.692
0.668
0.819
0.782
0.725
0.7
0.302
0.668
0.620
0.860
0.809
0.795
0.914
0.906
0.964
0.953
0.944
0.8
0.642
0.934
0.921
0.985
0.978
0.983
0.992
0.995
0.992
0.997
0.996
6
0.2
0.006
0.015
0.006
0.037
0.020
0.012
0.035
0.028
0.062
0.044
0.024
0.3
0.005
0.040
0.017
0.081
0.051
0.036
0.098
0.071
0.148
0.121
0.080
0.4
0.013
0.083
0.058
0.170
0.127
0.095
0.222
0.172
0.310
0.262
0.208
0.5
0.039
0.173
0.139
0.325
0.270
0.222
0.411
0.365
0.539
0.490
0.419
0.6
0.105
0.349
0.307
0.573
0.505
0.451
0.680
0.630
0.810
0.748
0.712
0.7
0.288
0.624
0.600
0.835
0.800
0.763
0.914
0.889
0.964
0.944
0.940
0.8
0.640
0.924
0.914
0.981
0.984
0.971
0.996
0.994
0.999
1.000
0.998
8
0.2
0.092
0.059
0.070
0.065
0.067
0.037
0.067
0.047
0.083
0.049
0.040
0.3
0.031
0.045
0.035
0.087
0.061
0.038
0.098
0.072
0.149
0.115
0.078
0.4
0.015
0.082
0.054
0.169
0.124
0.090
0.215
0.169
0.300
0.254
0.201
0.5
0.036
0.170
0.136
0.319
0.265
0.214
0.404
0.361
0.539
0.475
0.411
0.6
0.104
0.344
0.301
0.569
0.496
0.445
0.675
0.628
0.803
0.748
0.710
0.7
0.277
0.620
0.598
0.833
0.798
0.752
0.909
0.882
0.964
0.939
0.940
0.8
0.636
0.912
0.915
0.978
0.982
0.973
0.996
0.996
0.999
0.999
0.999
10
0.2
0.348
0.254
0.319
0.235
0.301
0.203
0.285
0.206
0.302
0.199
0.229
0.3
0.260
0.200
0.240
0.207
0.242
0.146
0.246
0.168
0.283
0.194
0.190
0.4
0.176
0.165
0.182
0.224
0.220
0.146
0.291
0.213
0.347
0.284
0.242
0.5
0.109
0.202
0.184
0.339
0.298
0.233
0.424
0.361
0.543
0.472
0.421
0.6
0.116
0.342
0.309
0.567
0.494
0.443
0.667
0.622
0.802
0.743
0.707
0.7
0.276
0.618
0.595
0.826
0.793
0.760
0.908
0.881
0.962
0.945
0.938
0.8
0.637
0.926
0.914
0.982
0.981
0.977
0.996
0.996
1.000
0.999
0.999
Page 61 of 70
-------
Table A7: Power when the lowest acceptable ratio 95% LCLmCPT/mCPT = 0.8
(Weibull distribution)
median
P5MR
Sample Size
21
22
23
24
25
26
27
28
29
30
2
0.2
0.097
0.070
0.135
0.097
0.180
0.137
0.108
0.181
0.135
0.223
0.3
0.185
0.168
0.243
0.231
0.318
0.294
0.215
0.357
0.253
0.406
0.4
0.332
0.335
0.413
0.405
0.490
0.483
0.374
0.555
0.436
0.617
0.5
0.505
0.542
0.605
0.623
0.686
0.693
0.580
0.765
0.646
0.801
0.6
0.720
0.757
0.801
0.823
0.860
0.860
0.815
0.909
0.851
0.934
0.7
0.898
0.925
0.920
0.937
0.950
0.939
0.940
0.934
0.940
0.960
0.8
0.858
0.898
0.816
0.848
0.867
0.810
0.857
0.760
0.805
0.836
4
0.2
0.077
0.047
0.117
0.069
0.159
0.112
0.084
0.141
0.110
0.174
0.3
0.190
0.145
0.261
0.192
0.337
0.260
0.224
0.333
0.280
0.393
0.4
0.373
0.319
0.477
0.397
0.578
0.480
0.462
0.581
0.517
0.654
0.5
0.621
0.566
0.735
0.667
0.816
0.747
0.736
0.825
0.788
0.870
0.6
0.856
0.838
0.919
0.890
0.958
0.931
0.935
0.966
0.956
0.977
0.7
0.983
0.978
0.993
0.987
0.997
0.994
0.996
0.998
1.000
0.999
0.8
0.994
0.996
0.989
0.996
0.993
0.991
0.997
0.986
0.992
0.992
6
0.2
0.070
0.043
0.104
0.065
0.151
0.100
0.075
0.130
0.100
0.167
0.3
0.176
0.131
0.245
0.178
0.323
0.243
0.204
0.316
0.254
0.376
0.4
0.347
0.297
0.457
0.387
0.554
0.468
0.426
0.561
0.493
0.639
0.5
0.592
0.544
0.701
0.652
0.796
0.733
0.697
0.821
0.757
0.868
0.6
0.837
0.824
0.904
0.879
0.948
0.934
0.916
0.964
0.947
0.979
0.7
0.981
0.977
0.993
0.988
0.996
0.996
0.995
0.999
0.999
0.999
0.8
1.000
0.999
0.999
1.000
1.000
1.000
1.000
0.998
0.999
0.998
8
0.2
0.073
0.054
0.110
0.076
0.144
0.107
0.075
0.134
0.095
0.166
0.3
0.169
0.129
0.239
0.177
0.316
0.238
0.196
0.309
0.252
0.367
0.4
0.342
0.292
0.447
0.375
0.547
0.458
0.424
0.554
0.491
0.634
0.5
0.582
0.535
0.703
0.646
0.790
0.731
0.691
0.820
0.758
0.867
0.6
0.833
0.823
0.908
0.881
0.947
0.928
0.914
0.965
0.944
0.978
0.7
0.974
0.975
0.990
0.988
0.995
0.995
0.994
0.999
0.999
0.999
0.8
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
10
0.2
0.222
0.235
0.225
0.245
0.266
0.251
0.178
0.276
0.208
0.297
0.3
0.243
0.231
0.298
0.262
0.364
0.305
0.242
0.372
0.294
0.415
0.4
0.370
0.329
0.462
0.401
0.557
0.474
0.423
0.566
0.491
0.636
0.5
0.585
0.541
0.700
0.643
0.790
0.726
0.686
0.813
0.751
0.862
0.6
0.827
0.825
0.903
0.875
0.947
0.927
0.918
0.965
0.947
0.978
0.7
0.978
0.977
0.991
0.986
0.995
0.995
0.995
1.000
0.999
1.000
0.8
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
The SAS Code used to generate the simulated data and the associated tables are presented in Appendix 14.2C.
Page 62 of 70
-------
14.2A Re-parameterization of Standard Weit , on
Given the definition of PDF and CDF from first principles:
(mCPT\K
P(mCPT,K,X) = 1 — e v x ) =0.5 (median)
(PSMRxmCPT\
Then:
P(P5MR x mCPT, k, X) = 1 - e~
(mCPT\K
e I x ) = 0.5 (median)
(P5MRXmCPT\K
> = 0.95 (5th percentile)
x
e ^
and
/mCPT\K
= ln(0.5)
V A
Divide (2) by (1), we have:
P5MR x mCPT
\ X
/P5MRxmCPT\K
—J = ln(0.95)
From (1):
mCPT
k = In
ln(0.95)
ln(0.5)
ln(0.95)
ln(0.5)
/ln(P5Mi?)
fmCPT\K , N
(—) = -ln(°'5)
(mCPT\
tcxlnl—-—J = ln[— ln(0.5)]
(mCPT\ 1
) = K '"["'nC0 5)]
1
In (mCPT) — ln(A) = — ln[— ln(0.5)]
K
l
ln(A) = In (mCPT) — — ln[— ln(0.5)]
K
l
= — [k In (mCPT) — ln[— ln(0.5)]]
K
l
= - [In{mCPTK)~ ln[— ln(0.5)]]
K
1
= - In
K
mCPTK
X = et
So...
k = In
— In
mCPTK 1
ln(o,5) J
ln(0.5)
ln(0.95)
ln(0.5)
mCPTK
/ ln(P5Mi?)
. X In
X = eK"*"l ln(0.5)J
(As shown in the main text)
= 0.05 (5th percentile)
(1)
(2)
(3)
(4)
Page 63 of 70
-------
14.2B Estimated Weibull Parameters of CPT data a tick repellency study
Background
In 2015, Buchel. K et. al published the results of a tick repellency study title "Repellent efficacy ofDEET, Icaridin,
and EBAAP against Ixodes ricinus and Ixodes scapularis nymphs (Acari, Ixodidae) In this study, there were 10
volunteers for each of 3 repellents x 2 tick species. Each volunteer tested 5 ticks every 30 minutes until CPT was
reached, up to 12.5 hours. The authors of the study kindly provided the raw data to EPA to allow us to investigate
the characteristics of tick CPT data (i.e. distributions and parameters of the distributions) to better develop a sample
size simulation.
Methods
After obtaining the raw data from the authors of study, EPA staff reviewed and made some corrections in the CPT
data per EPA definition of CPT.
Weibull distributions, normal distributions, and lognonnal distributions were used to analyze the data to determine
the best fit the CPT data. Weibull distributions were selected as the best fit distributions based on the lower AIC
values (Table 5).
The P5MRs of the distributions were calculated using the estimated parameters of CPT distributions, assuming the
data following Weibull distributions.
Conclusion
It is reasonable to assume that the CPT data of the tick study follow Weibull distributions
The estimated P5MR (5th percentile/median) of the tick CPT data ranges from 0.27 - 0.54
Table A8: compare the fitness of Weibull. nonnal. and lognonnal distributions for the CPT data
Species
product
AIC (smaller is better)
WEIBULL
NORMAL
LNORMAL
I. ricinus
DEET
9.526
35.761
11.215
EBAAP
22.620
43.660
25.601
Icaridin
6.838
47.639
9.157
I. scapularis
DEET
11.119
40.365
11.250
EBAAP
17.178
28.812
19.908
Icaridin
8.623
36.826
10.103
Note: yellow-shaded cells indicate the selected distributions
Table A9:
SAS code
Estimated Weibull Parameters using MLE
Species
product
WeibullScale
WeibullShape
p5
p50
P5MR
I. ricinus
DEET
4.276
3.773
1.946
3.880
0.502
EBAAP
3.821
1.981
0.853
3.175
0.269
Icaridin
8.792
4.219
4.348
8.060
0.539
I. scapularis
DEET
5.074
3.287
2.056
4.539
0.453
EBAAP
2.250
2.586
0.713
1.952
0.365
Icaridin
4.657
3.978
2.208
4.248
0.520
Page 64 of 70
-------
* Programmer: James Nguyen, USEPA
* Project: Tick Repellencv Studies
* Study: CPT data in Kerstin Buchel 2015 article
* Purpose: estimate parameters of the CPT data assume
* the da ta fo11ow Weibu11 di s tribu ti on
options Formdlim=,,=" nodate nonumber ls=100 ps=100;
Proc import datafile="F:\Insect Repellency\Tick Repellent Studies\PCT Kerstin Buchel 2015
data.xlsx"
dbms=xlsx out=Buchel replace;
run;
Proc sort data = Buchel; by species product; run;
ods graphics on;
ods rtf file="C:\Users\JNguyen\Desktop\Junks\Kaplan-Meier Survival Curves.rtf"
startpage=no;
proc lifetest data = Buchel method=km plots=(survival(atrisk=0 to 12 by 0.5));
time CPT*status(0);
strata Product;
by species;
run;
ods rtf close;
*===> testing distributions;
ods output FitStatistics=WEIBULL(rename=(Value=WEIBULL));
Proc lifereg data = Buchel;
by species product ;
model CPT*status(0)=/distribution=WEIBULL;
run;
ods output FitStatistics=NORMAL(rename=(Value=NORMAL));
Proc lifereg data = Buchel;
by species product ;
model CPT*status(0)=/distribution= NORMAL;
run;
ods output FitStatistics=LNORMAL(rename=(Value=LNORMAL));
Proc lifereg data = Buchel;
by species product ;
model CPT*status(0)=/distribution= LNORMAL;
run;
Data Distributions;
merge WEIBULL NORMAL LNORMAL;
by species product ;
if criterion = "AIC (smaller is better)";
run;
ods rtf file="C:\Users\JNguyen\Desktop\Junks\Kaplan-Meier Survival Curves.rtf"
startpage=no;
Proc print data = distributions noobs; run;
ods rtf close;
*===> Estimate Weibull parameters;
ods output ParameterEstimates=ParameterEstimates;
Proc lifereg data = Buchel;
by species product ;
model CPT*status(0)=;
run;
Proc transpose data = ParameterEstimates out=ParameterEstimates(drop=_NAME_);
where Parameter in ("Weibull Scale","Weibull Shape");
by species product ;
var Estimate;
Page 65 of 70
-------
ID Parameter;
run;
data ParameterEstimates;
set ParameterEstimates;
p5 = quantile('WEIBULL',
p50 = quantile('WEIBULL'
P5MR=p5/p50;
run;
ods rtf file="C:\Users\JNguyen\Desktop\Junks\Weibull Parameters.rtf" startpage=no;
Proc print data = ParameterEstimates noobs;
format Weibull_Shape Weibull_Scale p5 p50 p5mr 6.3;
run;
ods rtf close;
0. 05,Weibull_Shape,Weibull_Scale) ;
,0.S,Weibull_Shape,Weibull_Scale);
14.2C SAS Codes for Simulations
* Programmer: James Nguyen, USEPA *
* Project: Ticks Repellencv Studies *
* Purpose: Power sis/sample size calculation for *
* study i of 1 tick/15 rninuttes *
* Descrii.>1 i < hi:: *
* - 'i i i ll in l 'lis: Weibull T'1-rrrnl Lognorrnal, Uniform *
* - i M In i ograrns of th 'i nil utions *
* - I i 'i ires: PROC I I I I 'I I '1 mj PROC ICLI FETE ST *
* Date: 1/09/2018 *
options formdlim=,,=" ps=90 ls=90 nonumber nodate;
libname Ticks "C:\Users\JNguyen\Desktop\Ticks - 15 min interval";
%Macro distParam;
if upcase(Distribution) = "WEIBULL" then do;
* Weibull :::::: f (x, a , b) ;
a = log (log(0.95)/log(0.5))/log(P5MR) ; b = exp ( (1/a)*log(-
(MED**a)/log(0.5)));
end;
if upcase(Distribution) = "UNIFORM" then do;
* uniform U [ a F b] ;
a = MED* (0. 5*P5MR - 0.05) / 0.45; b = MED"4" 2 - a;
end;
if upcase(Distribution) = "NORMAL" then do;
* norm a 1 :::::: N (a , b) ;
a = MED; b = MED* (1-P5MR)/1.645;
end;
if upcase(Distribution) = "LOGNORMAL" then do;
"k log normal :::::: exp(N(a,b));
a = log(MED); b = (log(MED)-log(MED*P5MR))/l.645;
end;
%Mend;title;
%Macro generate',
if upcase(Distribution)
if upcase(Distribution)
if upcase(Distribution)
if upcase(Distribution)
%Mend;
"WEIBULL" then OPT = rand ( "Weibull", a, b) ;
"LOGNORMAL" then CPT = exp (rand ( "Normal", a, b) ) ;
"NORMAL" then CPT = rand("Normal", a, b);
"UNIFORM" then CPT = a + (b-a)*rand("Uniform") ,
%Macro CPT;
Page 66 of 70
-------
CPT=CPT* 60;
*==> the time to test 5 ticks is 20 minutes;
if CPT <= 3 then do; CPT = 0;
censor = 0; end;
else if CPT > &maxT*S0 + 3 then do; CPT = &maxT*S0; censor
= 1; end;
else do; CPT = 15*ceil((CPT-
3)/15); censor = 0; end;
CPT = CPT/60;
%Mend;title;
%Macro power;
ods select none;
ods output Quartiles=MPT;
Proc lifetest data = Simmer(keep=MED P5MR N Sim CPT Censor);
by MED P5MR N Sim;
time CPT*Censor (1);
run;
ods select default;
Proc datasets nolist; delete simmer; run;quit;
Data MPT;
set MPT;
if percent = 50;
power = (LowerLimit >= &K*Estimate);
run;
Proc SQL;
create table &dist&MED as
select MED, P5MR, N, avg(Power) as Power
from MPT
group by MED, P5MR, N;
quit;
%Mend;title;
%Macro PowerCPT(med=, P5MRS=, nmin=,nmax=,maxT=,K=,dist=,NSim=, seed=);
%let N=l;
%let P5MR&N = %nrbquote(%scan(&P5MRS,&N, %str( )));
%do %while (&&P5MR&N A=);
%let N=%eval(&N+1);
%let P5MR&N = %nrbquote(%scan(&P5MRS,&N, %str ( )));
%end;
%let N=%eval(&N-1);
fedo i = 1 %to &N;
%if &i = 1 %then %do; data All_&dist&MED; set _NULL_; run; %end;
Data Parameters;
MED = &MED;
P5MR = &&P5MR&i;
P5 = MED^P5MR;
label MED = "median" P5MR="5%-tile/median ratio";
run;
Data Parameters;
set Parameters;
Distribution = "&dist";
%dlstParam;
run;
data simmer;
Page 67 of 70
-------
call streaminit(Sseed);
set Parameters;
do N = &Nmin to SNmax;
do Sim = 1 to SNSim;
do ID = 1 to N;
end;
% generate;
output;
ID;
end;
end; *N;
drop a b;
run;
Data Simmer;
set Simmer;
%CPT;
run;
%power;
Data All_&dist&MED;
set All_&dist&MED Sdist&MED;
run;
Proc datasets nolist; delete Parameters MPT Sdist&MED; quit;
%end;
Data !'
run;
.Sdist._T15_M&MED._K%sysevalf(100*&K)_N&nmin._&nmax._D&maxt;
set All Sdist&MED;
Proc datasets nolist; save sasmacr; run;quit;
%Mend;
dm log ' 'PowmrCPT : • : 2, P5MRS=0.2
maxT=12, K = 0.6, dist= weibull, NSim=4000
dm log 'clear'; %PowerCPT(med=4, P5MRS=0.2
maxT=12, K = 0.6, dist= weibull, NSim=4000
dm log 'clear';%PowerCPT(med=6, P5MRS=0.2
maxT=12, K = 0.6, dist= weibull, NSim=4000
dm log 'clear';%PowerCPT(med=8,
maxT=12, K = 0.6, dist= weibull,
dm log 'clear';%PowerCPT(med=10,
maxT=12, K = 0.6, dist= weibull,
dm log 'clear'
maxT=12, K = C
dm log 'clear'
maxT=12, K = C
dm log 'clear'
maxT=12, K = C
dm log 'clear'
maxT=12, K = C
dm log 'clear'
maxT=12, K = C
dm log 'clear'
maxT=12, K = C
dm log 'clear'
maxT=12, K = C
dm log 'clear'
maxT=12, K = C
%PowerCPT{med=2,
7, dist= weibull,
PowerCPT (me d= 4,
7, dist= weibull,
PowerCPT (me d= 6,
7, dist= weibull,
PowerCPT ,
7, dist= weibull,
%PowerCPT(me d=10,
7, dist= weibull,
%PowerCPT(med=2,
B, dist= weibull,
PowerCPT (me d= 4,
B, dist= weibull,
PowerCPT (me d= 6,
B, dist= weibull,
P5MRS=0.
NSim=40C
P5MRS=0.
NSim=40C
P5MRS=
NSim=4
P5MRS=
NSim=4
P5MRS=
NSim=4
P5MRS=
NSim=4
P5MRS=
NSim=4
P5MRS=
NSim=4
P5MRS=
NSim=4
P5MRS=
NSim=4
0.3 0.
seed=
0.3 0.
seed=
0.3 0.4 0,1
seed=5613E
613E
613E
seed=5613E
0.3 0.4 0,1
seed=5613E
seed=
613E
seed=
0.3 0.4 0,1
seed=5613E
613E
0.3 0.4 0.£
seed=5613E
seed=5613E
0.3 0.
seed=
0.3 0.
seed=
seed=5613
613E
613E
,6 0.7 0.
,6 0.7 0.
,6 0.7 0.
0.6 0.7 0.
0.6 0.7 0.
0.6 0.7 0.
6 0.7 0.
6 0.7 0.
0.6 0.7 0.
0.6 0.7 0.
0.6 0.7 0.
6 0.7 0.
6 0.7 0.
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
nmin=10, nmax=20
Page 68 of 70
-------
dm log 'clear%PowerCPT(med=8, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.8, dist= weibull, NSim=4000, seed=56198);
dm log 'clear' ;%PowerCPT(med=10, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.8, dist= weibull, NSim=4000, seed=56198);
dm log ' 'PowmrCPT : • : 2, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.6, dist= weibull, NSim=4000, seed=56198);
dm log 'clear%PowerCPT(med=4, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.6, dist= weibull, NSim=4000, seed=56198);
dm log 'clear' ;%PowerCPT(med=6, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.6, dist= weibull, NSim=4000, seed=56198);
dm log 'clear' ;%PowerCPT(med=8, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.6, dist= weibull, NSim=4000, seed=56198);
dm log 'clear' ;%PowerCPT(med=10, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.6, dist= weibull, NSim=4000, seed=56198);
dm log 'clear%PowerCPT(med=2,
maxT=12, K = 0.7, dist= weibull,
dm log 'clear' ;%PowerCPT(med=4,
maxT=12, K = 0.7, dist= weibull,
dm log 'clear' ;%PowerCPT(med=6,
maxT=12, K = 0.7, dist= weibull,
P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
NSim=4000, seed=56138);
P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
NSim=4000, seed=56138);
P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
NSim=4000, seed=56138);
dm log 'clear%PowerCPT(med=8, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.7, dist= weibull, NSim=4000, seed=56198);
dm log 'clear' ;%PowerCPT(med=10, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.7, dist= weibull, NSim=4000, seed=56198);
dm log ' 'PowmrCPT : • : 2, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.8, dist= weibull, NSim=4000, seed=56198);
dm log 'clear' ;%PowerCPT(med=4, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.8, dist= weibull, NSim=4000, seed=56198);
dm log 'clear' ;%PowerCPT(med=6, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.8, dist= weibull, NSim=4000, seed=56198);
dm log 'clear' ;%PowerCPT(med=8, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.8, dist= weibull, NSim=4000, seed=56198);
dm log 'clear%PowerCPT(med=10, P5MRS=0.2 0.3 0.4 0.5 0.6 0.7 0.8
maxT=12, K = 0.8, dist= weibull, NSim=4000, seed=56198);
nmin=10
nmin=10
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
nmin=21
i.icmic: 1 i (11! i (:and Print Results;
libname Ticks "C:\Users\JNguyen\Desktop\Ticks - 15 min interval";
%let folder=C:\Users\JNguyen\Desktop\Ticks - 15 min interval;
%Macro SGPLOT(dist=, K=, nmin=, nmax=, maxt=);
title "Sdist median hours, K = O.&K";
Proc SGPLOT data = .dist._T15_M2_K%sysevalf(10*&K)_N&nmin._&nmax._D&maxt;
scatter x = Power/group = P5MR;
series x = ^ ower/group = P5MR;
refline 0.8 0,9/axis=y;
yaxis min=0 max=l;
run;
title "Sdist median = 4 hours, K = O.&K";
Proc SGPLOT data = Ticks.Sdist._T15_M4_K%sysevalf(10*SK)_NSnmin._Snmax._DSmaxt;
scatter x = N y = Power/group = P5MR;
series x = N y = Power/group = P5MR;
refline 0.8 0,9/axis=y;
yaxis min=0 max=l;
run;
title "Sdist median h hours, K = O.&K";
Proc SGPLOT data = .dist._T15_M6_K%sysevalf(10*&K)_N&nmin._&nmax._D&maxt;
scatter x = Power/group = P5MR;
Page 69 of 70
-------
series x = N y = Power/group = P5MR;
refline 0.8 £5„9/axis=y;
yaxis min=0 max=l;
run;
title "&dist median = 8 hours, K = O.&K";
Proc SGPLOT data = T i.cks . Sdist._T 15_M8_K%syseva 1 f (10*&K) _N&nmin._&nmax._D&maxt;
scatter x = N y = Power/group = P5MR;
series x = N y = Power/group = P5MR;
refline 0.8 £5„9/axis=y;
yaxis min=0 max=l;
run;
title "Scdist median id uours, K = O.&K";
Proc SGPLOT data = .dist._T15_M10_K%sysevalf(10*&K)_N&nmin._&nmax._D&maxt;
scatter x = Power/group = P5MR;
series x = N y = Power/group = P5MR;
refline 0.8 £5„9/axis=y;
yaxis min=0 max=l;
run;
%Mend;
%Macro print(dist=, K=, nmin=, nmax=, maxt=);
data &dist. K&K;
set Ti.cks . &dist._T 15_M2_K%syseva 1 f (10*&K) _N&nmin._&nmax._D&maxt
.&dist._T15_M4_K%sysevalf(10*&K)_N&nmin._&nmax._D&maxt
.&dist._T15_M6_K%sysevalf(10*&K)_N&nmin._&nmax._D&maxt
.&dist._T15_M8_K%sysevalf(10*&K)_N&nmin._&nmax._D&maxt
.&dist._T15_M10_K%sysevalf(10*&K)_N&nmin._&nmax._D&maxt;
run;
Proc transpose data = &dist._K&K out = &dist._K&K(drop=_NAME_);
by MED P5MR;
ID N;
var Power;
run;
title "&dist K=0.&K.0M;
Proc print data = &dist._K&K noobs label; format 6.3; run;
%mend;
feS'GPXOT(dist=Weibull, K = 6, nmin=10, nmax=20, maxt.=12) ,
tSGPLOT{dist=VJeihull, K=6, nmin=21, nmax=30, maxt=12) ,
tSGPLOT{dist=VJeihull, K=7, nmin=10, nmax=20, maxt=12) ,
tSGPLOT{dist=VJeihull, K=7, nmin=21, nmax=30, maxt=12) ,
tSGPLOT{dist=VJeihull, K=8, nmin=10, nmax=20, maxt=12) ,
feS'GPXOT(dist=Weibull, K=8, nmin=21, nmax=30, maxt.=12) ,
ods rtf file = "&folder\&dist 15 minutes.rtf" bodytitle;
%print (dist=Weibull, K= 6, nmin=10, nmax=2£5, maxt.=12)
%print (dist=Weibull, K= 6, nmin=21, nmax=30, maxt.=12)
%print (dist=Weibull, K=7, nmin=10, nmax=2£5, maxt.=12)
%print (dist=Weibull, K=7, nmin=21, nmax=30, maxt.=12)
%print (dist=Weibull, K=8, nmin=10, nmax=2£5, maxt.=12)
%print (dist=Weibull, K=8, nmin=21, nmax=30, maxt.=12)
ods rtf close;
Page 70 of 70
------- |