tfto sT4^ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
* A \ WASHINGTON D C., 20460
' e ¦
w
5 V^77 Jj? OFFICE OF CHEMICAL SAFETY
%, ¦> AND POLLUTION PREVENTION
March 18, 2023
MEMORANDUM
SUBJECT: Science and Ethics Review of a Protocol for Efficacy Testing of SCJ Personal
Repellent Products against Mosquitoes in a Field Study
FROM: Robert D. Mitchell III, Ph.D., Entomologist
Registration Division
Office of Pesticide Programs
Angela Myer, Ph.D., Entomologist/Biologist
Health Effects Division
Office of Pesticide Programs
Clara Fuentes, 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
THRU: Elizabeth Fertich, Acting Branch Chief
Registration Division
Office of Pesticide Programs
Kaitlin Saunders, Risk Manager
Registration Division
Office of Pesticide Programs
TO: Dan Rosenblatt, Acting Director
Registration Division
Office of Pesticide Programs
REF: S.C. Johnson & Son protocol for "Efficacy Testing of SCJ Personal Repellent
Products against Mosquitoes in a Field Study." Protocol ID: 90020741. February
2022.
Page 1 of 118
-------
S.C. Johnson & Son (SCJ) has submitted a protocol to the EPA for field testing of two skin-
applied repellents, one an aerosol product (EPA protocol number 4822PA51; DP 465257)
containing 10% p-menthane-3,8-diol (PMD) and the other a solid stick product (EPA protocol
number 4822PA52; DP 465253) containing 10% picaridin, against wild populations of mosquito
vectors, from both scientific and ethical perspectives. This review assesses the scientific aspects
of the proposed research for a product performance study to evaluate the efficacy of skin applied
insect repellents in terms of the recommendations of EPA Guideline OPPTS 810.3700: Insect
Repellents to be Applied to Human Skin/ Pesticide Product Performance Data Requirements for
Products Claiming Efficacy Against Certain Invertebrate Pests (hereafter referred to as the
"Product Performance Rule")2 and the Repellency Awareness Guidance For Skin-Applied Insect
Repellent Producers* concerning scientific merit of the proposed study. Ethical aspects of the
proposed research are assessed in terms of the standards defined by 40 CFR 26 subparts K and L
and the recommendations of the Human Studies Review Board (HSRB).
If the protocol is revised to address the recommendations outlined in this review, the resulting
research is likely to meet applicable scientific and ethical standards for research with human
subjects.
A. Completeness of Protocol Submission
The submitted protocol package, additional information on statistical justification for the
proposed sample size, and supplementary documentation of review by the WCG Institutional
Review Board (IRB) were reviewed for completeness against the required elements listed in 40
CFR §26.1125. The EPA's protocol review (Attachment 1) and ethics submission checklists
(Attachment 2) document how the submission will comply with the regulatory requirements for
protocols describing research with human subjects provided the EPA's recommendations are
addressed. The protocol submission will be complete under the criteria of the EPA's Human
Studies Rule when the Agency's recommendations outlined in the memo are addressed
adequately. However, the submission of a revised 'complete' protocol does not guarantee the
submitted data will be acceptable and support product label claims.
B. Summary Assessment of Science Aspects of the Proposed Research
The EPA summarized the protocol based on the information provided in the document submitted
by SCJ.
Note: The page number references in parentheses throughout the document refer to
specific pages of the submitted protocol.
1 EPA. Product Performance Test Guidelines; OPPTS 810.3700: Insect Repellents Applied to Human Skin. EPA
712-C-10-001. July 7, 2010. https://www.regulations.gov/document/EPA-HO-OPPT-2009-0150-Q011
2 EPA. 87 FR 22464: Pesticide Product Performance Data Requirements for Products Claiming Efficacy Against
Certain Invertebrate Pests. April 15, 2022. https://www.federalregister.gov/documents/2022/04/15/2022-
07963/pesticide-product-performance-data-reauirements-for-products-claiming-efficacv-against-certain#sectno-
reference-158.1756
3 EPA. Repellency Awareness Guidance For Skin-Applied Insect Repellent Producers. September 2013.
https://www.re gulations. gov/document/EPA-HO-QPP-2013 -0406-0003
Page 2 of 118
-------
Objectives
The primary objective of the protocol is to determine the duration of efficacy of two separate
personal repellent products (an aerosol spray containing 10% PMD as the active ingredient and a
solid stick containing 10% picaridin as the active ingredient) against wild mosquitoes that have
the potential to transmit pathogens of public health importance in a field setting using human test
subjects (p. 2). The human test subjects will be representative of the target populations for these
products, which is the typical user of skin-applied repellents against mosquitoes in the United
States.
Efficacy Endpoints
Efficacy will be measured as the duration between the time the test substance is applied and the
time the first confirmed mosquito landing on a test subject's treated appendage occurs, which
will be defined as the complete protection time (CPT). The first confirmed landing (FCL), which
is the endpoint of the study for each treated test subject, is defined as the time at which one
mosquito landing occurs and a second landing occurs within 30 minutes. The second landing
confirms the first landing. The median CPT (mCPT) across all the test subjects will be calculated
as the summary measure of CPT for each product tested (p. 2).
Overview and Test Species
The protocol will utilize human test subjects to determine the efficacy of two repellent products,
a 10%) PMD aerosol spray and a 10%> picaridin solid stick, against specific vectors encountered
at field test sites. The protocol specifies 14 treated test subjects per product per location (p. 25).
The registrant's rationale for using 14 treated test subjects (Attachment 3) instead of 13 treated
test subjects recommended by EPA (Attachment 4) is that:
First, it is usually desirable to have a sample size that is even to allow for an equal
number of male andfemale test subjects. Second, the 95% lower confidence limit of the
Kaplan-Meier median is the 3rd smallest value when the sample size is 14 but the 2nd
smallest value when the sample size is 13, so a sample size of 14 is likely to give a more
precise low-end estimate of the median. (Attachment 3)
The proposed testing will determine a mCPT for each product at two geographically distinct sites
where, "species from at least three genera Aedes, Anopheles, and Culex, will be targeted as
much as possible" (p. 17). The testing site in Wisconsin is designated as Eastern Temperate
Forest-Southeastern Wisconsin Till Plains and the Massachusetts site is designated as Eastern
Temperate Forest-Northeastern Coastal Zone. The following mosquito species have been
regularly collected at these two sites:
• Aedes cinereus
• Aedes stimulans
• Aedes trivittatus
• Aedes vexans
• Anopheles punctipennis
• Anopheles quadrimaculatus
• Coquilletidia perturbans
Page 3 of 118
-------
• Culex erraticus
• Culex pipiens
• Ochlerotatus canadensis
• Ochlerotatus excrucians
• Ochlerotatus fitchii
• Ochlerotatus japonicus
• Ochlerotatus thibaulti
• Ochlerotatus triseriatus
• Psorophora ferox
A standard application rate of 1.0 g/600 cm2 that is based on EPA's analysis of data derived from
dosimetry studies on aerosol spray products (Attachment 5) will be used for the PMD aerosol
spray (p. 16). Dosimetry testing will be conducted with the 10% picaridin stick at the laboratory
screening and training facility or at a designated field site prior to field efficacy evaluation to
determine the typical consumer rate to be used for efficacy testing (pp. 26-27).
Pre-Efficacy Testing Recruitment and Training
A pre-training day recruitment phase will employ several inclusion and exclusion criteria to
identify qualified candidates who are representative of a typical insect repellent-using population
from the general population of the United States. Potential test subjects will be selected based on
criteria listed in the protocol to ensure that subjects are qualified to participate in the study (pp.
3-4). To be included in the study a subject must: be able to prove that they are between 18-55
years of age with valid identification, read and understand English, not be an employee of SCJ or
an immediate family member, have reliable transportation to and from the test site and training
location, be willing to be exposed to and potentially bitten by mosquitoes, refrain from the use of
alcohol 12 hours before testing, refrain from nicotine and fragrance products like cologne during
testing, be willing to sign the informed consent document (ICD), be willing to have body
temperature checked as part of COVID protocols, and be a user of insect repellent products.
Candidates will be omitted from the study if they suffer from specific physical ailments that
would potentially endanger them during participation, such as asthma, skin problems, or
sensitivity or allergy to mosquito bites. Pregnant and lactating women will be excluded from
participation. Only candidates that provide consent by signing the ICD will participate in the
study. However, only those that subsequently demonstrate they are attractive to mosquitoes and
successfully complete aspirator training will be eligible to participate in field efficacy testing.
The selected candidates will be informed of the purpose of the study, the rules, and any potential
associated health risks and will be required to sign an ICD upon arrival at their designated
training session if they wish to proceed any further with the process. The laboratory or field site
screening and training phase (projected to take 2-3 hours) will include potential test subject
consent and selection, attractiveness testing, aspirator training, and dosimetry for the solid stick
picaridin product and will occur within two months of field efficacy trials. Lab-based
attractiveness training will use lab-reared adult female mosquitoes that have not been exposed to
human blood sources, while field-based attractiveness testing will rely on wild mosquito
populations (not lab-reared mosquitoes). However, subjects in the field training session will be
wearing bug suits provided by study staff to help prevent mosquito bites (pp. 23-24).
Page 4 of 118
-------
Lab-based training day procedures prior to efficacy testing will use lab-reared mosquitoes that
have not been exposed to human blood sources (p. 12), and the mosquitoes will be contained in
2' x 2' x 2' screened cages with two cloth sleeves (pp. 9-10). For lab attractiveness testing,
subjects will either insert their arm that is "protected with a sleeve that mosquitoes cannot bite
through" into a cage with 10 mosquitoes (p. 9) or will insert a covered arm that is "protected with
a sleeve the mosquito's proboscis cannot penetrate" into a cage with 50 mosquitoes (p. 23). A
subject is deemed attractive if five landings occur in one minute or less during the laboratory
attractiveness testing (p. 9). Subjects will have 3 attempts to confirm attractiveness before being
excluded from further participation (p. 23). Any subject determined to be unattractive to
mosquitoes (i.e., fewer than five landings within a one-minute exposure period after three tries)
will not be permitted to participate any further with the study and will be financially
compensated for their participation up to that point (pp. 9, 22). Subjects deemed to be attractive
to mosquitoes will be trained to identify when a mosquito has landed on their arm and practice
capturing them with an aspirator (p. 9). If aspirator training occurs in the lab, subjects will either
practice aspirating mosquitoes off the exposed skin of one forearm using 10 mosquitoes in a cage
(p. 9) or practice aspirating mosquitoes off their bug suit-covered arm using 50 mosquitoes in a
cage (pp. 9-10). A staff member will demonstrate the process first and then allow the subject to
practice until that person and the study staff "feels they are sufficiently proficient to participate
in the field test" (p. 10).
Field site training in lieu of laboratory training is included in the protocol as an option and is
projected to take 2-3 hours. This training will occur within 2 months of to the start of product
efficacy testing in the field. During attractiveness testing conducted in the field a subject will
wear a bug suit (along with gloves and closed-toe shoes), which will allow wild mosquitoes to
land on a subject without being able to bite the person wearing the suit. The bug suit is described
as a jacket and pants made of tightly woven breathable material (nylon) with a screened mesh
covering over the face that allows for air exchange but is impenetrable to mosquito bites (pp. 9,
27). Each person will keep their bug suit on for the entirety of attractiveness training so that
"they will not expose any skin directly to mosquitoes" (p. 9) and be considered attractive if there
are five mosquito landings on their suit-covered forearm or lower leg within a five-minute
exposure period. Subjects deemed to be attractive to mosquitoes will be trained to identify when
a mosquito has landed on their bug suit-covered forearm or lower leg and how to capture them
with an aspirator before they are able to bite or fly away. For landing identification and aspirator
training conducted in the field a subject will practice observing and aspirating mosquitoes that
have landed on their bug suit-covered limb until that person and the study staff feels they are
sufficiently proficient with the aspirator to participate in field efficacy testing.
Estimation of Skin Surface Area
Test subjects will have their non-dominant forearm and the opposite lower leg measured during
the laboratory or field training day to determine the appropriate treatment surface area, which
allows them to operate the aspirator with their dominant hand during testing (pp. 23-24). The
forearm will be measured at the wrist just below the wrist bone, at the crease of the elbow, and at
two points roughly equidistant between the wrist and elbow to establish the appropriate treatment
surface area as follows:
Page 5 of 118
-------
Forearm surface area (cm2) = CxD
C = Mean of the circumference measurements (in cm)
D = Distance between the wrist and the elbow circumference measurements (in cm)
The lower leg will be measured at the crease of the bent knee, immediately above the laterally
protruding ankle bone, and at two points roughly equidistant between the knee and ankle to
establish the appropriate treatment surface area as follows:
Lower leg surface area (cm2) = CxD
C = Mean of the circumference measurements (in cm)
D = Distance between the knee and the ankle circumference measurements (in cm)
Dose Determination (Dosimetry)
A dose determination test (dosimetry) for the 10% picaridin solid stick product will also be
conducted in the laboratory or in the field on the subject training day (pp. 26-27). Dosimetry is a
method used to determine a "typical consumer dose" of a topical repellent (OPPTS 810.3700,
Section (b), p. 8). During set-up, the Study Director will determine if the forearm or lower leg
will be used, and test subjects will be asked to wash their limb designated for treatment using a
gentle, unscented soap; dry the limb with paper towels; wipe the limb with a 70% solution of
isopropanol or ethanol in water; and allow it to dry (p. 26-27). Medical tape and bandaging will
be used to mark off the area of the limb to be treated, with the inner edge of the bandage/tape
aligned with the outer boundaries of the previously acquired limb surface area measurements (p.
27).
During the test, a test subject will be assigned a staff member for data recording (p. 27). The test
sample will be weighed on a tared scale, then subjects will be provided with the test sample and
proposed label directions for use and instructed to apply the product as they normally would to
achieve complete coverage (p. 27). After application, the test sample will be returned to the tared
scale and the amount applied recorded. Subjects will repeat the washing, drying, and ethanol
wipe procedures used during set-up. The treatment process will be repeated two additional times
per subject, alternating the treatment arm for each treatment (p. 10, p. 27). "No test subject will
be allowed to treat themselves at a rate greater than 2 g/600 cm2 " (p. 27). An average amount
applied will be calculated per subject, and the average rate for each test subject will be averaged
to determine an overall application rate (g/cm2) (p. 27).
EPA's standard aerosol dose of 1.0 g/600 cm2 will be used to apply the PMD aerosol (p. 23), so
dosimetry testing is not applicable to the aerosol product.
Calculation of Subject-Specific Treatment Amounts
The treatment amount (g) for each test subject will be calculated using the forearm and/or lower
leg surface area measurements as follows:
Page 6 of 118
-------
For the 10% picaridin stick:
Test substance (g) = Average application rate (g/cm2) derived from the picaridin stick dosimetry
determination multiplied by surface area of the subject's treatment area.
For the 10% PMD aerosol:
Test substance (g) =1.0 g/600 cm2 multiplied by surface area of the subject's treatment area.
Excel will be used to calculate these treatment amounts and the associated acceptable range
(±10%) of target amount), and numbers rounded to two decimal places or by default Excel
settings (p. 23).
Field Efficacy Testing
Field efficacy testing will occur at Brighton Dale Park in Kenosha County, Wisconsin and at
Hale Reservation in Westwood, Massachusetts (Norfolk County). The protocol also suggested
that alternative locations in the United States may be used if adequate mosquito density could not
be found at the two named locations (p. 17). Each product will be tested over the course of one
day at each test site. Testing may occur for 12+ hours (p. 8).
Study staff will use CO2 baited CDC traps at the Wisconsin testing site to collect and identify
mosquitoes on a weekly basis the month before field efficacy testing is conducted. Collection
and identification at the Massachusetts testing site will be conducted by the Norfolk County
Mosquito Control District using CO2 baited CDC traps and the data will be shared with SCJ
personnel (p. 17). The Study Director will utilize information provided on the United States
Geological Survey (USGS), Centers for Disease Control and Prevention (CDC), and state health
department websites to monitor any recorded occurrence of mosquito-borne diseases (pp. 20-21).
If the necessary information is not readily available on those websites, then the Study Director
will contact local health departments and mosquito control districts directly to acquire the data in
the county of the test location (p. 20).
Upon arrival at the designated test site, subjects will be requested to complete a COVID-19
screening, then an overview of the testing logistics will be provided. Subjects will be instructed
to wash a designated arm or leg with unscented soap, dry it with a paper towel, and then wipe
down the surface with a 70% isopropanol or ethanol solution and let it dry (p. 27). They will then
be provided a bug suit for protection. The suit is: "made of tightly woven nylon which allows air
exchange but is impenetrable to mosquito bites. The bottoms of the pant legs and ends of the
sleeves will fit snuggly around the wearer's boot (or high-top shoe) or wrist respectively; this
keeps mosquitoes out. The jacket has mesh on the sides (under arm) for better "breathability, " a
hood with mesh protecting the face area (a zipper allows removal of this mesh around the face)
and a drawstring band at the waist" (p. 27). Subjects will also be given gloves to protect their
hands. Aspirators will also be provided during set-up and subjects will be reminded on how to
use them (p. 27).
Bug suit sleeves or pant legs will be rolled up and secured with tape, and "the area of each
subject's designated limb (treated/untreated controls/alternates) that was previously measured
Page 7 of 118
-------
will [be] marked off with a medical tape wrapped bandage so that the inner edge of the
bandage/tape ran along the outer boundary of the measured area at both ends" (p. 28). Two
designated staff members will be used to apply the products to treated subjects (p. 28).
PMD aerosol applications: A standard application rate of 1.0 g/600 cm2 will be used for
the PMD aerosol product (p. 16). The test substance container will be placed on a scale,
and the scale tared. The test substance will then be sprayed from the container onto a test
subject's skin, then returned to the scale to determine the amount applied (p. 28). "If the
amount applied is below the required weight, more will be sprayed onto the skin targeting
the calculated amount (by weight) as closely as possible and then returned to the balance
to determine the amount applied" (p. 28). A staff member will then spread the product
using two fingers of a gloved hand to help ensure complete coverage. The glove used to
spread the product will be weighed before and after application to account for product
loss from the glove (p. 28).
Picaridin stick applications: Dosimetry, as described previously, will be conducted to
determine the picaridin stick product application rate prior to field efficacy testing (p. 16).
The solid stick product will be placed on a scale, and the scale tared. The test substance
will be swept over the treatment area, then returned to the scale to determine the amount
applied (p. 28). "If the amount applied is below the required weight, more will be applied
onto the skin targeting the calculated amount (by weight) as closely as possible and then
returned to the balance to determine the amount applied. The test substance will be
spread around the forearm directly from the stick application to ensure complete
coverage " (p. 28).
Four alternates will be available at each location at the start of the testing day to replace a test
subject if they fail to show up or drop out before testing begins, or if a test subject is
inadvertently treated with an amount of the test substance that is above the allowable maximum
(p. 25); see "Calculation of Subject-Specific Treatment Amounts" section above for details. The
alternate test subjects will be allowed to leave the testing site once the test substance has been
successfully applied to the treated test subjects (pp. 11-12). A 2-hour delay between application
and first exposure period will be used for the PMD aerosol testing, when "test subjects will
remain in a location that is environmentally similar to the test site (temperature, relative
humidity) but without the test system and out of direct sunlight" (p. 29). The picaridin stick
exposures will not be delayed (p. 29).
For the field efficacy test, the subjects will be reminded of how to respond to landing
mosquitoes, and then directed to the test area and separated by a minimum of 10 feet (p. 28).
Each 30-minute interval includes the following procedures:
At the direction of the study staff, all subjects will observe their treated limbs for a period
offive minutes. This period will be timed by a study staff member using a stopwatch and
test subjects will be instructed when to start and stop. Each subject will observe
themselves for landing mosquitoes on their exposed limb area. They will quickly attempt
to aspirate any landing mosquitoes from their skin before they bite. The test subject will
notify study staff when a mosquito lands on their skin so that study staff may write down
the time of the landing on a paper data sheet. Five-minute exposures will be made at 30-
Page 8 of 118
-------
minute intervals until a first confirmed land{ ing] occurs or up to sometime the Study
Director or Principal Investigator (PI) stops the test, due to inclement weather for
example, (p. 28)
Subjects will have access to the pavilion and be protected from the sun and mosquitoes in the
approximately 25 minutes following each 5-minute exposure. Following FCL subjects will be
instructed to return to the pavilion, remove tape/bandaging, and wash their treated limb to finish
their participation for the test day. All aspirated mosquitoes will be labeled with date and location
and saved for mosquito identification (p. 28).
Untreated control subjects (preferably one male and one female) "will remain in the field
continuously, along with the treated test subjects, but will only expose their untreated arm or leg
for a 5-minute period approximately every thirty minutes to confirm mosquito activity to reduce
the exposure to biting mosquitoes" (p. 26). These exposure periods will occur at the direction of
a staff member. The target mosquito landing pressure will be "a minimum average landing rate
of 5 lands per 5-minute exposure period" (p. 26), or "to have at least 5 lands in 5 minutes during
the test" (p. 29). During control subject exposures, "they will attempt to aspirate the landing
mosquitoes or shake the exposed limb before the mosquito has a chance to bite " (p. 29).
Untreated control subjects will cover their exposed limbs by rolling down their sleeve or pant leg
after 5 landings have occurred in a 5-minute exposure period. If the target landing pressure is not
met, "the Study Director or PI may allow further exposures of the untreated controls in the
expectation that adequate biting pressure will resume or decide to move the subjects to a nearby
location where biting pressure may be higher" (p. 29). The test will be stopped for all test
subjects if adequate landing pressure on the untreated control subjects is not met for 3
consecutive exposure periods (p. 29). This landing pressure assessment will be repeated once
approximately every 30 minutes "until all treated test subjects have experienced repellent
breakdown, or the Study Director or PI chooses to stop the test" (p. 11).
Sample Size and Number of Subjects
The protocol proposes to recruit 60-80 candidates (30-40 per site). At each site, 14 repellent-
treated subjects, 2 untreated controls, and 4 alternates will be used.
Randomization
Subjects will be identified by code numbers that will be assigned upon arrival to the training
session (p. 9). These codes will be used to randomly select 7 male and 7 female treated subjects
and 2 male and 2 female alternates per product on the day of efficacy tests (p. 26). The Study
Director will select untreated control subjects "from subjects proficient at aspirating landing
mosquitoes. This is to ensure that the subjects are well-versed in how to quickly recognize
landing behavior and remove those mosquitoes before they bite " (p. 26).
Margin of Exposure
Field testing of two skin-applied repellents, one an aerosol product containing 10% PMD and the
other a solid stick product containing 10% picaridin, will be conducted against wild populations
Page 9 of 118
-------
of representative mosquito vectors. A standard dose of 1.0 g/600 cm2 will be used for the PMD
product and dosimetry will be used to establish the appropriate application rate for the picaridin
product.
PMD: A 90-Day dermal study in rats (MRID 44438710) tested PMD (98.3 % pure) at increasing
doses, 0, 1,000 and 3,000 mg/kg/day. The No Observed Adverse Effect Level (NOAEL) = 1,000
mg/kg/day, and the Lowest Observed Adverse Effect Level (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 44438711) 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. Margins of
Exposure (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 subpopulations (infants and children) from the use of PMD in insect repellent
products applied to human skin.
Picaridin: The active ingredient picaridin is classified as toxicity category IV for Acute Oral
(LDso >5,000 mg/kg (MRIDs 51868905 and 44408748), Acute Dermal (LD50 >5,000 mg/kg
(MRID 51868907 and 44408749), Acute Inhalation (LC50 > 4.364 mg/L (could be waived based
on results from MRID 44408709), and Dermal Irritation (MRID 51868902). It is classified as
toxicity category III for Eye Irritation (i.e., caused moderate eye irritation that cleared in 7 days
or less; MRID 51868903) and is not a dermal sensitizer (MRIDs 51868906 and 44408752).
MOEs were not calculated because there are no endpoints of concern for the dermal route of
exposure. The entire chronic toxicity database was generated using dermal studies, including
developmental/reproductive and chronic studies, and there were no systemic toxic effects
identified that would be relevant to humans. In the acute dermal study, the LC50, No Observed
Effect Level (NOEL) and NOAEL were all greater than 2,000 mg/kg, and there was no evidence
of dermal irritation or dermal sensitization. The Agency concluded that there is reasonable
certainty of no harm to populations or subpopulation (infants and children) from the use of
picaridin in insect repellent products applied to human skin.
According to an EPA risk assessment for picaridin:4
Toxicity endpoints and points of departure were not selectedfor picaridin. A qualitative
human health risk assessment was conducted in lieu of a quantitative assessment, due to
the limited toxicity seen in the database. Although kidney effects were seen in rats in
dermal and oral toxicity studies, the effects are a result of accumulation of alpha 2-u
globulin (a2u-g). Since a2u-g is a male rat specific protein, kidney toxicity induced by its
accumulation is not expected to occur in female rats, mice, or in any other species,
including humans. Therefore, the kidney effects observed in male rats are not considered
appropriate for use in human health risk assessment. Outside of the kidney effects,
toxicity was limited to body weight changes seen only at the limit dose (1000 mg/kg/day);
4 Middleton, K., Figueroa, Z. December 5, 2014. Picaridin. Preliminary Human Health Risk Assessment in Support
of Registration Review. https://www.regulations.gov/document/EPA-HO-OPP-2014-0341-00Q7
Page 10 of 118
-------
human exposure at this level is not anticipated in conjunction with the use of products
containingpicaridin in accordance with registered labels, (p. 10)
Stopping Rules
The protocol includes the following conditions for stopping field testing:
• If adequate landing pressure on the untreated control subjects is not achieved during 3
consecutive exposure periods (p. 29).
• Inclement weather where test temperatures exceed 100°F or other unanticipated weather
conditions arise that make it unsafe to remain outdoors (p. 30).
• Adverse reactions are observed during testing where the Study Director decides to
remove the affected subject immediately and instructs them on obtaining appropriate
medical attention (p. 30).
• Safety reasons (p. 30).
• CPT is achieved by test subjects when the FCL occurs. Upon the FCL, a treated subject's
participation for the day is complete (p. 29).
• The Study Director or Principal Investigator chooses to stop the test (p. 11).
Withdrawal Criteria
Any subject that is shown to be unattractive to mosquitoes based on the criteria outlined in the
protocol will be prevented from participating in efficacy testing in the field (p. 9). Subjects may
withdraw from the study at any time without penalty or loss of benefits to which they may be
otherwise entitled (p. 5). However, if they do choose to withdraw early, they will only be paid for
the hours of their participation (p. 13). The Study Director or Principal Investigator may ask any
subject who does not follow instructions given in the ICD or by the study staff to withdraw from
the study or end a particular subject's participation in the study "at any time, andfor any reason"
(p. 22).
Criteria for Data Use from Withdrawn Subjects and Replacing Subjects During Field
Testing
No specific criteria were provided concerning data use from withdrawn subjects, and it was
proposed that alternates will be dismissed after product applications to treated subjects.
Data Collection
All data collected on subjects is confidential. Reckner (for the Wisconsin testing) and Focus on
Boston (for the Massachusetts testing) will collect data related to the general demographics of
repellent users in North America to obtain a pool of candidates for the study (p. 18). Data that
will be collected at each testing site includes:
1. CPT for each treated subject, which is the duration between test substance application
and the FCL on a treated subject's treated appendage (p. 16).
2. Landings data for untreated control subjects, described as "the time to reach 5 lands will
be recorded, or if 5 lands in 5 minutes is not reached the number of lands in 5 minutes
will be recorded' (p. 29).
Page 11 of 118
-------
3. Temperature, relative humidity, wind speed, wind direction, light intensity, and
precipitation will be collected by a weather station and/or other unspecified method (pp.
24, 29).
4. Mosquito identifications of specimens collected from subjects during testing (p. 17).
5. Mosquito-borne disease occurrences as reported on USGS and CDC websites, or
comparable information from local state health departments and Mosquito Control
Districts as necessary (pp. 21, 17).
All data will be stored by the SCJ Sensory and Claims Group (p. 4).
Statistical Analysis
Data will be analyzed by an external statistician (p. 4), with CPT data analyzed using the Kaplan-
Meier Estimator for survival data analysis (p. 29). Additional statistical methods like arithmetic
mean may be included if determined to be necessary based on the raw data (p. 29).
How and to What Will Human Subjects be Exposed
Subjects may be exposed to lab-reared mosquitoes that "will not have been exposed to human
blood sources" (p. 12) during lab-based training day procedures. If attractiveness testing occurs
in the field, subjects will either wear bug suits and gloves to "not expose any skin directly to
mosquitoes" (p. 9) or be exposed to a wild field population of mosquitoes when exposing their
skin for <5-minute exposure periods (p. 23). If aspirator training occurs in the field, subjects will
wear bug suits and gloves to "not expose any skin directly to mosquitoes'" (p. 10).
During the field efficacy studies, treated subjects will be exposed to a 10% picaridin solid stick
or a 10% PMD aerosol formula. The aerosol product containing 10% PMD will be applied at a
dose of 1.0 g of product/600 cm2 of skin to either a treated subject's arm or lower leg. Subjects
could also be exposed to the solid stick picaridin formulation during dosimetry testing (p. 27).
Field exposure time for each subject may be 12+ hours for a single day of product performance
testing. In the field, all subjects' uncovered limb will be exposed to wild natural mosquito
populations relevant to public health where pathogens have not been detected at the test site for
at least one month prior to testing.
Good Laboratory Practice (GLP) Compliance and Quality Assurance
Compliance with 40 CFR Part 160 (Good Laboratory Practice Standards) was reported as such:
A final report will be issued by the Study Director and will include all the required
elements to comply with 40CFR 160.185 and the most recent version of SOP GLP-RPT-
01. The Quality Assurance Unit will inspect at least one phase to ensure the integrity of
the study. In addition, the final report will be reviewed against the protocol, SOPs, and
raw data for accuracy. A statement will be included in the report specifying inspection
dates, phases, and the dates findings were reported to the Study Director and
Management, (pp. 30-31)
Compliance with FIFRA and EPA regulations
Page 12 of 118
-------
Data resulting from execution of this protocol as well as study conduct will be reviewed
by the EPA for compliance with FIFRA 12(a)(2)(P) and 40 CFR 26 subparts K, L, and M, and
will be independently audited by a QAU for compliance with Good Laboratory Practice
Regulations (40 CFR 160). The QA representative will conduct critical phase inspections to
ensure study integrity and maintain written and signed records of each inspection. The EPA will
consult with the HSRB about the protocol and the EPA's review as required under 40 CFR 26.
Study Site Location and Testing Facility
Field Study Locations:
• Wisconsin site - Brighton Dale Links Golf Club, 830 248th Avenue, Kansasville, WI 53139
• Massachusetts Site - Hale Reservation, 80 Carby Street, Westwood, MA 02090
• Additional unnamed sites
Laboratory and Field Screening and Training Sites: No specific location information provided
Principal Investigator: Daniel Usry, S.C. Johnson & Son
Study Sponsor: S.C. Johnson & Son, Inc., 1525 Howe Street, Racine, WI 53403
C. Compliance with Applicable Scientific Standards
This protocol should comply with applicable scientific standards after EPA's comments and
recommendations listed in the review section below are addressed.
D. EPA Science Comments and Recommendations
EPA recommends the study protocol should be revised as follows:
1. Picaridin stick references
Remove all references to the 10% picaridin stick if the protocol will not be used to
evaluate that product. The Agency was initially asked to review a protocol that would be
used for testing a 10% PMD aerosol repellent (EPA registration number 4822PA51; DP
465257) and a 10% picaridin solid stick repellent (EPA registration number 4822PA52; DP
465253) against mosquitoes. The Agency was informed via email notification on
11/4/2022 that SCJ would no longer pursue the picaridin solid stick testing. The protocol
was evaluated with both products listed even though SCJ does not wish to pursue picaridin
stick testing at this time. If there is the possibility that the protocol will be used in the
future to assess the efficacy of the 10% picaridin stick, then references to that product
should remain in the protocol. However, a clear distinction must be established between
experimental procedures intended for evaluation of the 10% PMD aerosol versus the 10%
picaridin stick, including product-specific dosimetry and application parameters. If the
picaridin stick testing will not be conducted at the same time as the testing of the PMD
Page 13 of 118
-------
aerosol product, the Agency recommends contacting the EPA to verify that this protocol is
still valid to assess efficacy of the 10% picaridin stick product before testing begins.
2. Field site selection and composition
a. The protocol's selected field sites and list of common species found at those sites
(p. 17) do not include Aedes aegypti or Aedes albopictus as required under 40 CFR
158 for products making a claim against mosquitoes. Absence of one of these
species during testing will render the data generated unacceptable to satisfy the
representative species requirement to support any mosquito claims. Under the
Product Performance Rule,5 testing on the following genera and species is required:
• Anopheles (,Anopheles albimanus or Anopheles jreeborni or Anopheles
gambiae or Anopheles hermsi or Anopheles punctipennis or Anopheles
quadrimaculatus or Anopheles stephensi); and
• Aedes {Aedes albopictus or Aedes aegypti); and
• Culex {Culexpipiens or Culex quinquefasciatus or Culex tarsalis).
The Agency requires the presence of these representative species (as listed above)
between sites (minimum of 2), but all 3 do not have to be present at each testing
site.
The mosquito species specified above represent taxa that are major disease vectors
(e.g., Ae. albopictus/Ae. aegypti for Zika virus, Cx. pipiens/quinquefasciatus for
West Nile Virus, etc.). Testing on these major disease vectors is required to support
any claims against mosquitoes, which are considered invertebrate pests of
significant public health importance under the Product Performance Rule. However,
inclusion of these required test species does not guarantee the submitted data will be
acceptable and support product label claims.
b. Revise the protocol to include the expected mosquito prevalence and specify that
the predominant mosquito species will differ between test sites to validate that they
are distinct habitats supporting different predominant species. OPPTS 810.3700
(Section (k), p. 27) notes that field tests for mosquito repellency should be
conducted in at least two distinct habitats where the predominant mosquito species
differ. The Agency also recommends that the sites be separated by a geographic
distance greater than the flight distance of the mosquito species present at the sites
to confirm the independence of selected test sites.6
5 EPA. Rule; 87 FR 22464: Pesticide Product Performance Data Requirements for Products Claiming Efficacy
Against Certain Invertebrate Pests. April 15, 2022. https://www.federalregister.gov/documents/2022/04/15/2022-
07963/pesticide-product-performance-data-reauirements-for-products-claiming-efficacv-against-certain#sectno-
reference-158.1756
6 Fuentes, C., Myer, A. September 29, 2022. Science Review of a Study Report on the Field Efficacy Test of an Oil
of Lemon Eucalyptus and Methyl Nonyl Ketone-Based Repellent Spray Against Mosquitoes.
https://www.epa.gov/svstem/files/documents/2Q22-
10/3d.%20EPA%20Science%20Review%20Mimikai%20mosauito%20studv%20MIM006%209-29-2022.pdf
Page 14 of 118
-------
3. General study design
The overall study design described in the document should be revised to provide detail to
all proposed procedures. The chronological order of procedures is also unclear in both the
protocol and ICD. The following format is suggested to describe study procedures in the
following chronological order:
A. Recruitment
B. Informed Consent
C. Attractiveness Testing (in the laboratory)
D. Aspirator Training (in the laboratory)
E. Limb Measurements (in the laboratory)
F. Dosimetry (in the laboratory)
G. Calculation of Subject-Specific Treatment Amounts
H. Field Site Monitoring
I. Field Testing
Furthermore, the document includes an abridged overview of the protocol in the beginning
of the document (pp. 1-6), followed by the ICD (Appendix A of the document, pp. 7-15),
and then another section of protocol procedures (Appendix B of the document, pp. 16-33).
To improve readability and flow, EPA recommends that the document be reorganized to
include all protocol procedures (described in a cohesive, chronological order) as the main
text, and the ICD and other relevant documentation included as appendices (see comment
#17, part (a) below). A table of contents at the beginning of the document is also
recommended. All procedures should be consistent between the ICD and the study
protocol
4. Laboratory versus field training/testing
The Agency recommends that all subject procedures preceding efficacy test days
(including attractiveness testing, aspirator training, limb measurements, and dosimetry
tests) be conducted only in the laboratory and NOT in the field due to potential exposure
to pathogen-infected wild mosquitoes and uncomfortable weather conditions. The
laboratory also provides a more controlled setting where environmental conditions are
consistent, the origin and status of lab-reared mosquitoes are known, and the use of a bug
suits is unnecessary. There would be no need to expose subjects to field conditions
(including wild mosquitoes and potentially uncomfortable environmental conditions) if
they are deemed unattractive to mosquitoes during laboratory-based attractiveness testing.
a. Specify in the protocol that the Final Report will include pertinent information
about the mosquitoes used during all laboratory procedures such as age,
developmental stage, methods of sexing, time of last blood meal (if relevant), and
whether and for how long they were starved pre-testing. The OPPTS 810.3700
Guideline (Section (j), p. 25), recommends that test mosquitoes should be fed a
10% sucrose solution but no blood meal before a test and starved for 12-24 hours
immediately prior to the test.
Page 15 of 118
-------
b. Mosquitoes used in all laboratory procedures should be lab-reared, certified
pathogen-free, female mosquitoes. Revise the protocol to indicate how the
pathogen-free status and sex will be confirmed and documented. Molecular testing
is typically performed to ensure pathogen-free status of lab-reared colonies.
Mosquitoes not having been exposed to a human blood source (p. 12) does not
necessarily assure they are pathogen-free.
c. Revise the protocol to clarify that mosquitoes will be disposed of after use with a
subject and will not be used by more than one subject. The Agency recommends
changing out the mosquitoes between attempts if a subject undergoes multiple
attempts during aspirator training and recommends replacing used mosquitoes with
unused ones between attempts if a subject is not attractive. Mosquitoes should also
not be re-used between attractiveness testing and aspirator training.
d. Describe procedures for killing mosquitoes after each use (e.g., placed in vials with
>70% ethanol and/or placed in a freezer for 24 hours) before disposal.
5. Attractiveness testing (laboratory)
a. Review the attractiveness testing criteria, OPPTS 810.3700 Guideline (Section (j),
p. 26), and provide detailed justification if departing from the Guideline
recommendations.
b. Clarify if all subjects will be tested once or more than once for attractiveness. It is
stated that subjects will get up to 3 attempts to demonstrate attractiveness, but not
specifically stated that additional attempts will not be made once a subject is
determined to be attractive (p. 23).
c. The Agency requests that raw data from attractiveness testing, which includes
number of attempts and time of landings, be recorded and reported in addition to the
other data that will be collected from the study (additional details under comment
#17 below).
d. Provide rationale for the mosquito density in the test cages during attractiveness
testing (p. 9), which deviates from the OPPTS 810.3700 Guideline
recommendation. The proposed size of the cage is consistent with the OPPTS
810.3700 Guideline (Section (j), p. 25), but the recommended insect density in the
Guideline is 200 mosquitoes per cage which is equivalent to 1 mosquito per 1,160
cm3. The current description in the protocol is the equivalent of 1 mosquito per
4,640 cm3 and no explanation is provided as to why a lower mosquito density will
be used. A justification for any deviation from the Guideline should be provided.
e. Remove this statement in section 8.5.2 of the protocol (p. 23): "The subjects' arm
will be protected with a sleeve the mosquito's proboscis cannot penetrate and the
hand will be protected with a glove. " This section should be revised to specify that
the skin of the one inserted forearm will be exposed (not covered) to mosquitoes for
the attractiveness tests, as recommended in OPPTS 810.3700, Section (j), p. 26.
6. Aspirator training (laboratory)
a. Describe the specific criteria required to determine when subjects are sufficiently
proficient to aspirate mosquitoes after landings and participate in field tests. The
Page 16 of 118
-------
protocol says that subjects will be deemed proficient when a "study staff member
feels they are sufficiently proficient' (p. 10). Determine a stopping point in the
training where proficiency has been met for each subject based on a required
number of successful attempts out of a total number of replicate attempts. Indicate a
distinct cutoff time limit for each replicate attempt.
b. Explicitly state the number of mosquitoes per cage used for aspirator training. For
aspirator training procedures, the protocol mentions 10 mosquitoes per cage on
page 9 and 50 mosquitoes per cage on page 10.
c. The use of a bug suit jacket on the arm used for observing landings during aspirator
training in the lab is not recommended without further description/justification of
the bug suit (p. 10).
7. Limb measurements (laboratory)
a. Specify that a flexible measuring tape should be used to determine the equidistant
points between the ankle and knee, and the equidistant points between the wrist and
elbow used for circumference measurements on each subject. This will reduce the
variability in limb surface area estimates between subjects.
b. The amount (g) of each product needed on the forearm and lower leg of each
subject to achieve the target application rates should be determined before efficacy
test day(s).
8. Dosimetry (laboratory)
a. Dosimetry testing should occur in the laboratory prior to efficacy test days.
Dosimetry tests cannot immediately precede field testing since the time required for
analyses of the doses derived from dosimetry and calculation of the test dose for the
solid stick product would significantly impact testing time in the field.
b. Revise the protocol to include more specific information about dosimetry testing,
including the number of subjects that are required to derive a typical consumer
dose, and when the testing will be conducted in relation to other training exercises.
See the relevant sections of the OPPTS 810.3700 Guideline for additional guidance
(Section (i), pp. 23-25). The revised protocol should include the following:
i. Confirm dosimetry test is conducted on each subject's forearm
(alternating between the limbs and conducting 3 separate treatments
preceded by washing/drying/ethanol wipe procedures).
ii. Describe units measured and to what decimal place. The Agency
recommends the use of a scale that measures grams up to 3 decimal
places.
iii. Specify that 25 subjects are to be used for dosimetry testing, in
accordance with EPA's science comments in a previous repellent
protocol review.7
7 Fuentes, C., Bohnenblust, E., Arling, M. March 30, 2018. Science and Ethics Review of a Protocol for Laboratory
Evaluation of Skin Applied Tick Repellent Product Containing OLE.
Page 17 of 118
-------
iv. Specify that the weighing method used to determine the amount applied
during each treatment will be performed by a staff member.
v. Provide the specific gravity of the stick product.
vi. Specify that the "typical consumer dose" estimated from the study will
be reported as both weight (mg per cm2) and volume (mL per cm2).
vii. Specify the calculations that will be used to scale the standard dose to
the surface area of each subject's limb or limbs.
viii. Specify that the mean dose applied by each subject (average application
dose of individual subjects) and the grand mean of doses applied across
all subjects will be calculated.
c. The rationale for the dose rate used in efficacy testing should be documented in the
Final Report.
d. Based on previous reviews and feedback from the HSRB, EPA recommends that
subjects be instructed to review a draft product label and apply the product, rather
than being told to "apply the test repellent to their own limb as they would normally
apply a repellent to achieve complete coverage" (p. 27). The draft label directions
for use should explicitly state how the product should be applied, not just "apply
evenly."
e. Remove the statement that "No test subject will be allowed to treat themselves with
a greater than 2 g/600 cm2 application rate" (p. 27). This is not possible to achieve
during an application. It is inappropriate to limit the application during the
dosimetry phase where subjects are observed making their "typical" application by
following the product label directions.
f. The Agency recommends that SCJ consult with the EPA after establishing the
typical consumer dose prior to initiating field testing.
g. Confirm that the "test sample" in dosimetry testing (p. 27) is the picaridin product
proposed for field efficacy testing to support label claims. A confidential statement
of formula (CSF) and certificate of analysis (CoA) should be submitted to the
Agency to confirm that each batch of the substance tested is the proposed product.
h. Specify that dosimetry testing will be performed with the final stick product in a
container that is identical to what the commercial packaging will be.
i. Confirm that EPA's standard aerosol dose of 1.0 g/600 cm2 will be used to apply
the PMD aerosol (p. 23) on efficacy test days, so dosimetry testing is not applicable
to the aerosol product.
9. Procedures on efficacy testing days (field sites)
a. Describe the bug suit in detail, including specifications such as type of fabric, the
density of the fibers, mesh size, how it prevents mosquito bites, and specific
references to other studies (if available) in which it was used successfully for a
similar purpose. EPA does not have sufficient information regarding the proposed
suit to evaluate the appropriateness of its use in this study. OPPTS 810.3700
https ://www.epa. gov/sites/default/files/2018-
04/documents/la. ole tick repellent protocol science and ethics review final 3-30-18.pdf
Page 18 of 118
-------
(Section (k), p. 28) specifies that "with the exception of the treated area, subjects'
heads, trunks, and limbs should all be covered with light-colored material through
which insects cannot bite. " For a bug suit to be considered for use by the Agency,
the suit should be: (1) impenetrable to mosquito bites, (2) be breathable enough to
avoid heat stress to subjects, (3) be able to be rolled up and secured with tape, and
(4) light-colored. The protocol and ICD should be revised to specify what clothing
subjects will be instructed to wear on field test days. The EPA determined in past
decisions that Tyvek suits made of high-density polyethylene fibers (plastic) were
unacceptable when proposed for a similar purpose.8
b. Specify a maximum duration for testing from repellent application to the end of the
last exposure period, without the "+" symbol. The ICD (p. 8) notes, "testing on
each test day may last up to 12+ hours, plus the time to review instructions, set up
the test site, and cover and apply the repellent to your arm or leg." This statement
does not provide a discrete test duration and the meaning of "cover" is unclear, so
this statement should be re-written with precise wording. It is recommended that
SCJ staff set up the test site and pavilion prior to the arrival of subjects.
c. The Agency recommends the use of the terms "subject(s)," "treated subjects(s),"
"untreated control subject(s) or "control subject(s)," and "alternate subject(s)." All
iterations of "subject" should be clearly defined when first discussed in the protocol
and used consistently throughout. The protocol uses the phrase "test subject(s)" to
refer to all subjects or treated subjects in different parts of the document and is
therefore confusing. OPPTS 810.3700 (Section (b), p. 6) states that a "human
sub ject is a living individual about whom an investigator conducting research
obtains either data through intervention or interaction with the individual or
identifiable private information. By this definition, both untreated control subjects
and treated subjects are considered human subjects of repellent efficacy testing. "
Individuals should be referred to as "candidates" until consent has been given.
d. Provide a detailed account of proposed stepwise, chronological procedures that will
occur for each 30-minute testing interval. At the start of field-testing exposure
periods, the untreated control subjects should establish landing pressure by
exposing their treated limb for <5 minutes (for 5 landings or 5 minutes, whichever
occurs first) immediately preceding treated subjects exposing their treated limbs for
<5 minutes (until a landing is observed or for 5 minutes, whichever occurs first).
Each subsequent 30-minute interval should also start with untreated control subjects
exposing their untreated limb (for 5 landings or 5 minutes, whichever occurs first),
immediately followed by treated subjects exposing their treated limb for <5 minutes
(until a landing is observed or for 5 minutes, whichever occurs first). Specify that
subjects will be resting under the pavilion during the times when their 5-minute
exposures are not occuring within the 30-minute intervals. Procedures should be
described to prevent cross-contamination between subjects and detail all tasks to be
performed by staff members during each 30-minute interval.
8 Fuentes, C., Hull-Sanders, H., Arling, M. December 18, 2020. Science and Ethics Review of a Protocol for Field
Evaluation of Two Topically Applied Insect Repellent Products Containing IR3535.
https://www.epa.gov/sites/default/files/2021-
01/documents/a. livful ir3535 protocol science and ethics review 2020 12 18.pdf
Page 19 of 118
-------
e. Specify that all mosquitoes that land on subjects' exposed skin will be collected for
species identification/counts and pathogen screening, but those that only land on
clothing (or other protective gear) of subjects will not be collected.
f. Clarify if two products are simultaneously tested on the same day at each test site,
or if the picaridin stick testing will occur at a different time from PMD aerosol
testing using the same (or different) sites. If the two products are tested on the same
day, then the distance between the exposure areas used for aerosol versus stick
testing should be reported.
g. A confidential statement of formula (CSF) and certificate of analysis (CoA) should
be submitted to the Agency to confirm that each batch of the substance(s) tested on
treated subjects is/are the proposed product(s).
h. Specify that the pavilion will be enclosed (screened-in) to exclude mosquitoes. The
protocol indicates that a pavilion will be available for subjects on efficacy test days
(p. 21), but it is unclear if the pavilion provides a rest area where subjects can leave
the treated limb uncovered/undisturbed while not being exposed to mosquitoes.
i. EPA recommends that the treated subjects are paired with another treated subject or
a staff member during exposures to help ensure accurate recording of CPT data and
removal of landing mosquitoes. OPPTS 810.3700 (Section (k), p. 29) states:
"Subjects may work in pairs to assist each other to identify landings and collecting
insects. Because clustering of subjects may confound results, at least 3 m/10 ft
should be maintained between pairs. "
j. Specify how many staff members will be assigned to treated subject pairs (if
applicable), and what task(s) will be performed by each member assigned to treated
subjects.
k. Specify if each untreated control subject is paired with the other or if each untreated
control is paired with a staff member (specify number) during exposures for each
product/site. The protocol only states, "A member of the study staff will assist the
control subjects during their exposures" (p. 29). Specify what task(s) will be
performed by each staff member assigned to untreated control subjects.
1. Shaking off mosquitoes during any of the exposure periods is not recommended for
the purposes of collecting specimens for identification and pathogen screening in
field studies of mosquito repellency testing. The protocol section discussing
untreated control subject exposures notes that "they [unspecified] will attempt to
aspirate the landing mosquitoes or shake the exposed limb before the mosquito has
a chance to bite" (p. 29). The protocol should be revised to remove reference to
shaking off mosquitoes, and to specify who will aspirate mosquitoes off the
exposed skin in this context.
m. Revise the protocol discussion around the minimum mosquito landing rate. The
Agency's criterion is at least 5 landings within 5 minutes on each of the untreated
control subjects for each 30-minute interval, not an average of 5 landings over
multiple time points or across both subjects. Remove from the protocol the
minimum average landing rate of 5 lands over 5 minutes (p. 26); If there is not
adequate pressure, testing on that day must be cancelled and rescheduled (see
Agency Recommendations regarding stopping rules below (comment #15)).
n. Revise the second sentence in protocol section 18.3.9 (p. 29) to clarify procedures
for if/when the target landing pressure is not met. The target landing pressure
Page 20 of 118
-------
should be reworded to align with the Agency's criteria (described in the previous
bullet). Furthermore, the phrase ''further exposure of the untreated controls'' is
imprecise and could inappropriately imply that the untreated controls will be
exposed for more than 5 minutes within a 30-minute interval. These statements
should be revised to state: "The target landing pressure is > 5 mosquitoes landing
within 5 minutes on each of the untreated control subjects for each 30-minute
interval. If this standard is not met, the Study Director may allow testing to continue
in that location with the expectation that adequate landing pressure will resume or
decide to move the subjects to a nearby location where biting pressure may be
higher until the stopping criteria is reached."
o. The Agency recommends supplying a proposed efficacy testing schedule, as a chart
or informative graphic (or both), to clarify several important factors including the
proposed and alternate test location(s), time of day testing is expected to occur, and
which specific product(s) will be tested on each test day.
p. Clarify what events need to occur before subjects are directed to exposure areas,
q. Avoid delaying exposures following application. During this time no CPT data will
be collected, and potential CPT time points will be unknown. For this reason, it is
not recommended to prolong delaying first field exposure unless the registrant
establishes criterion for ensuring that CPT recorded after a prolonged first exposure
delay is accurate and unlikely to have occurred at an earlier time point.9 EPA
recommends that when exposures are delayed, the delay period should be
established to ensure that a minimum of three consecutive exposure periods occur
before subjects experience an FCL. If the protocol retains the delay between
product application and exposure and an FCL occurs within the first three exposure
periods, the CPT will be counted as 0 hours. The Study Director may not replace a
subject who experiences an FCL.
r. Revise the restrictions language throughout the protocol/ICD. The wording is
unclear concerning all restrictions applying both before and during the test and
should be modified to reflect the following recommendations. OPPTS 810.3700
(Section (c) p. 9) specifies that, "Subjects should avoid alcohol, tobacco, and
scented products (perfume, cologne, hair spray, lotion, soap, etc.) for at least
twelve hours before and throughout the test. " Furthermore, the EPA recommends
that test days be spaced apart by a minimum of 72 hours if the same subjects are
used on multiple test days.10
s. Specify that subjects will be contacted and reminded of the restrictions above 48
hours prior to test day(s) and at the start of test day(s). The method in which
subjects will be contacted (phone, e-mail, etc.) should be specified, and the script
provided.
9 Fuentes, C., Hull-Sanders, H., Arling, M. March 25, 2021. Science and Ethics Review of a Protocol for Field
Evaluation of Skin Applied Mosquito Repellent Product Containing Oil of Lemon Eucalyptus and Methyl Nonyl
Ketone, https://www.epa.gov/sites/default/files/2021-
04/documents/lc. epa science ethics review memo w att mimikai mosquito mim-006 3-25-21.pdf
10 Fuentes, C., Bohnenblust, E., Arling, M. March 30, 2018. Science and Ethics Review of a Protocol for Laboratory
Evaluation of Skin Applied Tick Repellent Product Containing OLE. https://www.epa. gov/sites/default/files/2018-
04/documents/la. ole tick repellent protocol science and ethics review final 3-30-18.pdf
Page 21 of 118
-------
t. Clarify what will determine use of either the forearm or lower leg during efficacy
testing. The limb specified for treatment should be consistent across all subjects
within a test day/site. If lower legs are used, the use of the right versus left leg
should be randomized between subjects,
u. Confirm the number of subjects that will be treated on test days and revise the ICD
to reflect the correct number of subjects (p. 12).
v. Specify that a portable weather monitoring device will be used on site during
efficacy test days to record conditions specific to the test areas on efficacy test days
(see comment #17 below for more details). The protocol lists "weather station" as
study equipment "for measuring temperature, relative % humidity and wind
speed/direction during test" (p. 24). However, it is unclear whether this is a portable
or permanent weather station,
w. Confirm that the untreated control subjects will return to the pavilion, remove
tape/bandage, and wash the exposure limb after they have completed exposures for
the day to align with the end-of-day procedures described for treated subjects.
Field site monitoring
a. The Agency recommends pairing CO2 baited CDC miniature light traps and BG-
Sentinel traps for a more accurate estimate of species presence and include device-
specific parameter details like the amount of CO2 that is released, and specify any
lures used for monitoring. CO2 baited CDC miniature light traps are not the most
efficient tool for monitoring Aedes mosquitoes (particularly Ae. albopictus and Ae.
aegypti, i.e., container-breeding species) and could underestimate their population
in the testing area. Using appropriate and well-defined trapping methods should be
taken into consideration for confirming that the mosquito species required by the
Product Performance Rule are captured during site monitoring.
b. Revise the protocol and ICD to clarify site selection, species identification, and
disease screening procedures to be used during the site monitoring period. The
following revisions are recommended to align with OPPTS 810.3700 (Section (k),
subpart (2), p. 27):
i. Specify that all mosquitoes collected before and during efficacy tests
will be screened for West Nile Virus (WNV) and all other mosquito-
borne pathogens relevant to human health that could be transmitted by
the species located at the test sites. The pathogens to be screened for at
all sites should be stated in the protocol.
ii. Specify the appropriate molecular method(s) to be used for pathogen
screening, and that any pathogens identified as a result of the screening
will be reported to subjects and included in the Final Report.
iii. Specify that the mosquito specimens collected will be identified by
genus and species, and if possible, by subspecies or strain. The number
in each taxon collected at each time point should be reported. Ensure
that a taxonomically trained staff member is performing the
identifications.
c. The EPA has recommended in previous reviews that the selected field sites be
certified free from mosquito-borne pathogens for at least 4 weeks within 25 miles of
Page 22 of 118
-------
the test site, and that this monitoring data be collected immediately preceding the
test date(s).
d. Revise the protocol to include additional details about mosquito monitoring at the
test sites including trap type (including manufacturer/model information), number
of traps, trap locations within the site (and how they will be situated in relation to
exposure areas), mosquito species targeted, when (dates and time of day) traps will
be active and intended frequency of specimen collection. The trapping procedures
should be consistent between sites. The protocol should indicate that the Final
Report will contain trapping procedure details and include the monitoring data that
was collected (see comment #17 below).
e. If mosquito control district data alone will be used for site monitoring prior to
efficacy tests, then detailed mosquito control district SOPs of the applicable
site(s)/district(s) should be provided in the revised protocol to clarify surveillance
methods in a manner that addresses all details discussed in parts (a) through (d)
above.
11. Test substance application on efficacy test days
a. Specify whether the arm or leg will be chosen for applications at each test day (or
earlier), and the criteria for determination of treating the arm or leg.
b. Revise the protocol to provide more detail describing how staff members will be
allocated to specific tasks associated with application procedures, and what
precautions will be implemented to ensure the test substance is not disrupted after it
is applied in the correct amount (± 10%) to a treated subject.
c. Specify that the areas to be treated correspond to the surface area measurements (on
the forearm from the wrist just below the wrist bone to the elbow crease; on the
lower leg immediately above the laterally protruding ankle bone to the knee crease).
d. Revise the protocol to specify that a calibration procedure (modified from OPPTS
810.3500 Premises Treatments, Section (d), Option #3, p. 7) is performed prior to
aerosol product applications on efficacy test days, as described below:
i. To determine the quantity sprayed per second, 5 beakers will be
sprayed with the aerosol product for 3 seconds each. The product
container will be weighed before and after each spray and the
difference recorded. The mean value of the 5 replicates will be
determined and that result divided by 3 to determine the average
amount of product applied per second of spraying (g/s). Based on this
determined rate and subject limb measurements, the duration (seconds)
of spraying needed to achieve the standard dose (1.0 g/600 cm2) on
each treated subject will be calculated.
ii. The identical application device in which the product will be marketed
(as similar to the final product container as possible) should be used for
calibration and efficacy testing to most accurately emulate end-use.
Standardization of the application device, spray distance, and staff used
between repellent calibration procedures and subject repellent
applications is recommended to ensure accurate and precise application
of the standard aerosol dose.
Page 23 of 118
-------
e. The Agency recommends revising the protocol to specify the following application
methods for each product (if both products are to remain in the protocol), because
applications directly from the end-use container/applicator and subsequent
weighing on a scale (p. 28) are inaccurate for targeting specific doses, unless a
standardized container/spray distance/staff members are used across applications
and appropriate calibration procedures are performed.
i. To achieve the standard dose (1.0 g/600 cm2) for the aerosol product:
SCJ staff will apply the product onto each treated subject using the
spray duration (seconds) determined from calibration (described in part
(d) above). Following applications, the product should be spread across
the limb sections demarcated by tape/bandaging using gloved finger(s).
ii. To achieve the target dose (determined by dosimetry) for the solid stick
product: SCJ staff will determine subject-specific application amounts
(in grams) based on limb surface area measurement prior to efficacy
testing. During applications, a staff member will use a spatula and
weigh boat to weigh out the calculated subject-specific application
amount and use gloved finger(s) to apply and spread the repellent onto
the limb sections demarcated by tape/bandaging on treated subjects.
f. The ICD (p. 11) states: "The repellent will be applied to your skin using either a
pipette or a spatula or similar device, or it may be sprayedfrom a container. A staff
member will use two glovedfingers the repellent will be spread on your skin. " The
ICD should be revised to specify the application procedures recommended by the
Agency in part (e) above.
g. Specify that the standard dose for aerosol products (1.0 g/600 cm2) is recommended
by the Agency for skin-applied repellent testing based on an analysis of dosimetry
studies previously reviewed by the EPA and HSRB (Attachment 5).
h. Ensure that the test substance is applied almost simultaneously and consistently to
all treated subjects to minimize deviations. Please describe what procedure will be
followed to ensure consistency across applications (see parts (d) and (e) above).
i. Clarify where the repellent will be applied on test days (e.g., under the pavilion, in a
designated protected area in the field, in the laboratory, etc.).
Randomization
a. Revise the randomization process (pp. 9-11) to include more precise details.
Additional information should describe when the code numbers will be assigned to
study candidates/subjects and what method will be used to choose the code numbers
at random when assigning subjects to the treated, untreated control, and alternates
group (e.g., whether they be computationally assigned, drawn from a hat, etc.). The
Agency recommends that code numbers are assigned to subjects on their training
day in the laboratory (or earlier) before any training or testing occurs, and those
numbers are retained for the duration of the study.
b. Untreated control subjects should be selected randomly, rather than "selected by the
Study Director from subjects proficient at aspirating landing mosquitoes'' (p. 26).
Only subjects that passed the aspirator training should be present at the field site
Page 24 of 118
-------
during efficacy testing, and therefore all subjects should have an equal chance of
being assigned as an untreated control subject.
c. Clarify whether the same subjects will be allowed to participate in more than one
day of testing, and if both products will be tested on the same day(s). If different
subjects are used between products and/or sites/days, then the randomization
procedure for assigning subjects to each product and/or site/days should be
specified. The protocol should be revised to indicate that the subject assignments
(14 treated, 2 untreated controls, and 4 alternates) refer to assignments per product
per site. A single subject should only be counted once in determining the sample
size per product per site. To avoid potential cross-contamination of two products
containing different active ingredients, the same subjects should not be used to test
both products on the same day/at the same site. If the same subjects are to be used
on multiple test days, test days should be spaced apart by a minimum of 72 hours.
d. The protocol notes that 30-40 candidates will be recruited per site (p. 3). Clarify if
this number recruited is for testing one or both products.
13. Statistical analysis
a. The Agency recommends replacing the text under "Statistics" on page 29 of the
protocol with the following: "CPT data will be analyzed using the Kaplan-Meier
Estimator for survival data analysis, and the medians CPT (mCPT) and their 95%
CIs [confidence intervals] (where log-log transformation is applied to survival
function to obtain the confidence interval) will be estimated and presented along
with figures of Kaplan-Meier survival curves." The log-log transformation should
be used to calculate the 95% CI since the sample size of this study was obtained
from EPA's simulations of power vs. sample size where the log-log transformation
was used to calculate the 95% CI.
b. Specify that landing data collected from untreated control subjects will solely be
used to assess landing pressure during efficacy testing, and that untreated control
data will not be compared statistically to data collected from the treated subjects.
14. Withdrawn subject data
The ICD notes that alternates will be excused from testing and allowed to leave the test
site once the last treated subject has received their application, and no one has chosen to
withdraw at that point (p. 12). The protocol should specify that data from withdrawn
subjects should be treated as right censored if the subjects are not replaced. An mCPT
cannot be determined or unreliably estimated if there are right-censored data due to
withdrawals, especially withdrawals that occur during earlier testing intervals. Therefore,
careful experimental design and planning should be used to help ensure that an adequate
number of subjects provide FCLs. OPPTS 810.3700 states:
When testing a repellent with a long period of effectiveness, it is critical to the
reliability of the data to minimize "right-censoring" of the data—i.e., the number
of subjects who do not experience a failure of efficacy before the end of the test.
Withdrawal of test subjects from the study before failure of efficacy decreases the
Page 25 of 118
-------
sample size and may compromise the validity or utility of test results. (Section (k),
p. 28)
15. Stopping rules
a. The Agency recommends expanding the stopping conditions of the protocol (p. 30) to
include scenarios of mosquito landing pressure falling below acceptable levels and
bad weather, based on a previous EPA review.11 Specifically, testing should cease if:
i. 15% of all exposure periods (projected based on the maximum test
duration time) are skipped or missed due to inadequate landing
pressure (<5 mosquito landings within 5 minutes on each of the
untreated control subjects).
ii. 15% of all exposure periods (projected based the maximum test
duration time) are skipped or missed due to bad weather.
iii. Three consecutive periods are missed due to inadequate landing
pressure (<5 mosquito landings within 5 minutes on each of the
untreated control subjects).
iv. Three consecutive periods are missed due to bad weather.
b. Specify a maximum duration time for testing without the "+" symbol (See comment
#9, part (b) above for further detail) and include this duration as a stopping rule.
c. The Agency recommends defining "bad weather" conditions in the stop rules that
may compromise the test results (i.e., rain that results in fewer mosquitoes out
foraging). OPPTS 810.3700 (Section (k), p. 27-29) specifies that testing should not be
conducted or continued if wind speed exceeds 16 kph/10 mph. Weather conditions
may be checked in advance of the test day(s) to avoid scheduling testing during bad
weather.
16. References
a. The EPA recommends removing any references to the WHO publication Guidelines
for Efficacy Testing of Mosquito Repellents for Human Skin.12 The Kaplan-Meier
calculation presented in Annex 3 (Estimation of Median and Confidence Interval of
Complete Protection Time Using the Kaplan-Meier Survivor Function) of this 2009
WHO publication is incorrect. In particular, it appears that the WHO misinterpreted a
formula13 and therefore used incorrect values in the calculation which resulted in a
wrong answer in the WHO example.
b. Provide an appropriate citation for statements regarding ecoregion designations of the
proposed field sites (p. 17).
c. Provide an appropriate citation for the parenthetical reference for "2020 Neilson" (p.
18).
11 Fuentes, C., Bohnenblust, E., Arling, M. June 29, 2017. Science and Ethics Review of a Protocol for Field
Evaluation of Three Topically Applied Insect Repellent Products Containing IR3535.
https://www.epa.gov/sites/default/files/2017-07/documents/ir3535 protocol science and ethics review 6-29-
17.pdf
12 Link at: http://whalibdoc.who.int/hq/2009/WHO HTM NTD WHOPES 2009.4 eng.pdf
13 More specifically, it seems that they selected the wrong "n" to use.
Page 26 of 118
-------
Data collection and reporting
a. The Agency recommends inclusion of sample data collection sheets as part of the
protocol, and all raw datasheets should be included with the final report:
i. Landings data for treated subjects on efficacy test days - The following
should be recorded in tabulated form per treated subject (by code number): (1)
time (hours:minutes) when product was applied, (2) start time of each field
exposure period, (3) time of first landing, (4) time of FCL, and (5) CPT
calculated as the duration of time (hours:minutes) from time of product
application to time of FCL. The product tested, test date, and site location
should also be specified.
ii. Landings data for untreated control subjects on efficacy test days - The
following should be recorded in tabulated form per untreated control subject (by
code number): (1) start time (hours:minutes) of each field exposure period, (2)
time of each mosquito landing (hours:minutes), (3) total number of mosquitoes
that landed in 5 minutes or less (when 5 landings happen in less than 5 minutes).
The product tested, test date, and site location should also be specified.
iii. Dose applications on efficacy test days - The following should be recorded in
tabulated form per subject (by code number): (1) nominal amount of product
applied (grams of product per 600 cm2 of skin), (2) subject sex (male or
female), (3) limb used for applications (forearm or lower leg), (4) right or left
limb used for application, (5) calculated surface area of treated limb, (6) and
time of application (hours:minutes). The product tested, test date, and site
location should also be specified.
iv. Actual dose applied on test days - The following should be recorded in tabulated
form per treated subject (by code number) and provided as supplemental
information in an appendix: (1) nominal amount applied (g); (2) product loss from
glove (g) = glove weight after application - glove weight before application; (3)
actual amount applied (g) = nominal amount - product loss from glove; and (4)
actual rate of application (g/cnr) = actual amount applied (g) ^ skin surface area
treated (cm2). The product tested, test date, and site location should also be
specified.
v. Weather monitoring on efficacy test days - Specify that weather conditions
during testing (including temperature, relative humidity, cloud cover,
precipitation, light intensity, and wind speed) will be monitored periodically
throughout the study and reported, as recommended in OPPTS 810.3700
(Section (k), p. 31). Any deviations from these weather reporting procedures
should be justified.
vi. Mosquito monitoring prior to efficacy testing - The following should be
recorded for mosquito samples collected from each site: (1) site name, (2) site
county and state, (3) trap type (CDC light trap or BG sentinel trap), (4) trap
location/designation, (5) date of collection, (6) time of collection, (7) mosquito
species in sample, (8) number of specimens within sample by species (count
data). This data can be reported with the molecular-based pathogen screening
results.
Page 27 of 118
-------
vii. Mosquitoes aspirated during efficacy tests - Mosquito species and number of
specimens (by species) collected should be reported separately for each subject.
The product tested, test date, and site location should also be specified.
viii. Attractiveness testing - The following should be recorded in tabulated form per
subject (by code number): (1) Duration of < 1-minute exposure period
(seconds), (2) number of landings within the < 1-minute exposure period or
seconds passed to reach 5 landings. If applicable, data from multiple exposures
should be provided (with a maximum of 3 attempts per subject).
ix. Aspirator training - The following should be recorded in tabulated form per
subject (by code number): (1) exposure period duration (a time limitation should
be specified, e.g., 5 minutes to aspirate 5 mosquitoes successfully before they
bite), (2) number of successful capture attempts of the mosquitoes after they
landed and before they bit (i.e., 5 captured out of 5 landings = 100%), (3) data
from multiple exposures if applicable (a limit should be specified, e.g., no more
than 3 replicates). Consider using a 'smash test' to test for blood-feeding, if
applicable, i.e., taking the collected mosquitoes, immobilizing through chilling
or another immobilizing method, then physically crushing each mosquito on a
piece of white paper to determine blood-fed status; if blood is observed on paper
during the test, then the aspiration attempt should be considered unsuccessful.
x. Limb measurement - The limb measurements described in the protocol (p. 23-
24) should be recorded and reported for each subject in tabulated form.
xi. Calibration of aerosol spraying procedures - Data for calibration procedures
recommended in comment #11 (Section (d), subsection (i)) should be recorded
and reported for each staff member performing the calibration.
xii. Treatment amount calculations - The calculations described in the protocol
for determining target application amounts (g), and acceptable minimums and
maximums of (±10% of target) should be reported for each subject (by code
number), including surface area values, applicable application rates, and all
calculated values.
xiii. Dosimetry study - The following should be recorded in tabulated form per
subject per application (by code number): (1) pre-application
container/applicator weight, (2) post-application container/applicator weight, (3)
replicate number of each application (OPPTS 810.3700, Section (i), p. 24
specifies a minimum of 3 replicate applications per subject but consultation with
a statistician regarding potential issues of statistical power is generally
recommended), (4) the amount applied by weight, (5) the amount applied by
volume, (6) calculations used to determine the application rate for each replicate
made by each subject, (7) calculation of mean application dose of individual
subjects. The grand mean of doses applied across all subjects should also be
reported.
b. For all training sessions/dosimetry tests/field efficacy tests, specify how many
subjects will be assigned to each staff member recording data or paired with other
subjects in a manner that ensures accurate data recording and successful conduct of
proposed procedures (e.g., staff member assigned to each person, or subjects paired
with each other and monitored by staff).
Page 28 of 118
-------
c. For field monitoring and efficacy tests, specify how collected mosquitoes will be
stored (labeled vials marked with identifying information such as date, site, time of
collection, subject ID number, etc.) and transported to the laboratory for species
identification and counting (by species).
d. The Agency recommends that if a single landing on a treated subject during an
exposure period is followed by a missed exposure period (either through inadequate
landing pressure or bad weather) then the first landing will be treated as a confirmed
landing. If a confirmed landing occurs during an exposure preceded by a missed
exposure period (either through inadequate landing pressure or bad weather) then
CPT will be recorded as the earliest time point in that preceding delay.14
e. The word "land" instead of "landing" is occasionally used throughout document.
Revise throughout so the term "landing" is used in references to data collection.
f. Specify procedures to ensure that the time of application is accurately recorded for
each treated subject. Accurate recording of application times is critical to estimates of
CPT.
g. Confirm two products are proposed for testing in this protocol (cover letter on p. 34
refers to four products used in field testing).
h. Specify that all reported numbers will follow significant figures rules (appropriate
significant digits in measured values, and significant digits carried over in
calculations and rounded answers).
i. Specify what data will be collected by Reckner and Focus (p. 4 under "Data
Analysis" heading), and what data will be collected by SCJ staff. On page 4, it is not
clear if these independent recruitment agencies will be collecting data solely related
to recruitment.
18. QA/QC plan
Revise the protocol to specify procedures providing for periodic quality assurance
inspections adequate to ensure the integrity of the study and consistency with the
requirements of EPA's Good Laboratory Practices regulations (40 CFR part 160)
(OPPTS 810. 3700 (Section (a), p. 12). The protocol statement of "The Quality
Assurance Unit will inspect at least one phase to ensure the integrity of the study" is
vague and should be clarified.
19. Final report
Refer to the OPPTS 810.3700 Guideline (Section (g), pp. 23-24) when drafting a Final
Report for the study and ensure that the report includes details specified in the Guideline
as well as the required elements established under 40 CFR 160.185. Good Laboratory
Practice Standards (GLP) as defined in 40 CFR part 160 apply to both laboratory and
field studies of repellent efficacy (OPPTS 810.3700, Section (a), p. 6), and the Final
Report submission must be accompanied by the information required under 40 CFR
14 Fuentes, C., Bohnenblust, E., Arling, M. June 29, 2017. Science and Ethics Review of a Protocol for Field
Evaluation of Three Topically Applied Insect Repellent Products Containing IR3535.
https://www.epa.gov/sites/default/files/2017-07/documents/ir3535 protocol science and ethics review 6-29-
17.pdf
Page 29 of 118
-------
26.1303. The conduct of the proposed study should follow GLP standards as described in
40 CFR part 160, and all differences between the practices required by this part and those
used in the completed study should be detailed in the GLP compliance statement of the
Final Report. All deviations from the final EPA/IRB approved protocol should also be
reported in detail and provided with a justification. The Agency recommends that the
main body of the study report provide a detailed narrative of how the study (or studies)
were conducted from subject recruitment through data analysis to avoid delays in the
review process.
E. Summary Assessment of Ethical Aspects of the Proposed Research
Here is a summary of the EPA's observations about the ethical aspects of the proposed research.
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
of two topically-applied mosquito repellents (10% Icaridin Stick and 10% PMD Aerosol)
through testing the product against mosquitoes in the field using human subjects. As
intended, the data resulting from this proposed study will be used to support registration
of the two products being tested. Efficacy at preventing mosquitoes from landing on each
subject will be expressed as CPT, which is defined as the time between application of the
repellent product and the occurrence of the first mosquito landing on the treated skin
followed by a second landing within 30 minutes. These data will be combined and analyzed
to determine a median CPT (mCPT), which will be used to develop product labeling. The
research has societal value because people are at risk of contracting mosquito-borne
diseases, and such risks can be mitigated by the use of insect repellent products.
In addition, the protocol calls for establishing a typical consumer dose for the stick
product containing picaridin. These data are necessary as there have not been studies to
establish standard dosing for this type of product, and a standard dose is necessary in
order to conduct field testing. The typical consumer dose derived from the dosimetry
phase of the study will be used as the dose for the picaridin stick product during the field
testing of the product's efficacy.
2. Subject Selection: The protocol calls for testing each product with 14 treated subjects,
with an equal number of males and females (p. 3). There are discrepancies between the
protocol and consent form, where the consent form notes that only 12 treated subjects
will be needed for each test day (p. 8). An additional two individuals will participate in
the testing as untreated control subjects, monitoring mosquito landing pressure
immediately prior to each exposure period. Untreated controls will be selected by the
Study Director based on their assessment of the subjects' proficiency at aspirating
mosquitoes. In addition, four individuals will be enrolled as alternate subjects to take the
place of any subjects who withdraw before or on the day of testing (at least two alternate
subjects of each gender). A total of 20 individuals (14 treated subjects, 2 untreated
control subjects, and 4 alternate subjects) will be selected to test each product. Therefore,
a total of 40 subjects would be needed assuming each individual participates only in a
single test day. The protocol calls for recruiting 30-40 individuals per test site, for a total
Page 30 of 118
-------
of 60-80 individuals recruited.
The protocol calls for recruiting a diverse pool of candidates. The pool will be balanced
by gender. Ages will range from 18-55 years old, with approximately 20%-40% in the
18-30 range, 30%-50% in the 31-44 range, and 30%-50% in the 45-55 range (p. 3).
Racially, the goal is to recruit a pool including individuals identifying as: Caucasian
(50%-85%), African American, Asian, or other (5%-25%), and Hispanic (5%-25%) (p.
3).
Subjects will be recruited from the area local to the test sites (Kansasville, Wisconsin and
Boston, Massachusetts) by an independent recruitment agency at each site (Reckner in
Wisconsin, Focus on Boston in Massachusetts) (pp. 4 and 18). The protocol includes
demographic targets for recruitment (p. 18). Enrollment will be limited to English-
speaking individuals (p. 18). In addition, subjects must not be employees of SCJ or
immediate family members of SCJ.
The recruitment materials provided with the protocol indicate that first contact with
candidates will be made by the recruitment firm (Procedure for Initial Contact in
Recruiting Test Subjects via email for Insect Repellent Bio-Efficacy Studies). This
contact script describes the purpose of the study; the procedures that will be used for
training, attractiveness testing, and efficacy testing; compensation for subjects; and
eligibility criteria. The end of this document that would be emailed invites candidates
who self-identify as meeting the criteria to review the informed consent form, to ask
questions if they have any, and to sign the informed consent form if they are interested in
participating (Procedure for Initial Contact in Recruiting Test Subjects via email for
Insect Repellent Bio-Efficacy Studies, p. 3). The protocol notes that candidates who are
interested in participating in the study and who meet the necessary criteria would be
invited to attend a subject training session led by study staff members (pp. 17-18).
Prior to field testing, subjects will participate in a mosquito attractiveness test and
training on how to use an aspirator. There are two proposed methods to verify
attractiveness to mosquitoes - in the lab using mosquitoes in a cage and in the field using
wild mosquitoes (p. 23). In the lab, testing would involve placing an untreated forearm
into a cage measuring 2'x 2' x 2' and containing either 10 or 50 adult female mosquitoes
(pp. 9-10). Subjects will be considered attractive if they have five mosquito landings on
their arm within one minute. Field-based attractiveness testing would involve subjects
wearing a full bug suit, exposing a forearm or lower leg, and counting mosquito landings.
In the field, a subject would be deemed attractive if they receive five mosquito landings
on the untreated skin within five minutes. The protocol notes that subjects will have three
chances to demonstrate attractiveness to mosquitoes, but if they do not acquire at least
five landings in the prescribed period, they will be deemed unattractive and ineligible to
continue in the study.
The protocol also notes that"subjects will be taught to use a battery-powered aspirator
and will practice using the aspirator to collect landing mosquitoes until they and the
study stafffeels [sic] they are sufficiently proficient to participate in the field tesf (p. 23).
Page 31 of 118
-------
On each day subjects are exposed to the test substance or mosquitoes, female subjects
will be required to take a pregnancy test to confirm their eligibility to participate in the
research (p. 4). This test will be administered by the subject alone in a private bathroom
at the start of the test day. A female member of the study staff will discuss the results of
the pregnancy test with the subject in a private setting. If the subject is interested in
continuing to participate in the study, the female staff member will verify the negative
test result privately (p. 4).
Subjects will also be required to adhere to pre-testing restrictions in order to participate in
the testing. Subjects will be instructed not to drink alcohol for the 12 hours prior to the
test, and to refrain from smoking, chewing tobacco, and using fragrance products during
the test day (p. 12).
3. Risks to Subjects: The protocol discusses potential hazards associated with participating
in this study including: 1) adverse reaction to the test substances, 2) exposure to biting
mosquitoes, 3) exposure to mosquito-vectored diseases, 4) general risks of being in the
field, 5) risk of exposure to Covid-19, and 6) unanticipated loss of confidentiality.
The protocol notes that risks will be minimized as follows. To mitigate risk of adverse
reaction to the test substances, safety data sheets will provide evidence of low risk when
used as directed (p. 20). The dose for the picaridin stick product will be established by
dose determination testing, and the protocol calls for ensuring that the dose is no more
than 2 grams/600 cm2 (p. 17). During the dosimetry testing, subjects will make the
application and immediately wash the limb to which the application was made. Following
the three applications during the dosimetry phase, subjects will have washed each of the
limbs to which applications were made to remove the product applied. To mitigate the
risk of exposure to biting mosquitoes during the field testing, individuals with known
allergies to mosquito bites will be excluded, subjects will be trained on aspirating
mosquitoes before they bite, subjects will wear a bug suit, and subjects will only expose
one forearm or lower leg periodically (p. 20).
To minimize the risk of contracting any mosquito-borne diseases, the protocol notes that
subjects will be informed about the symptoms of mosquito-borne illnesses and instructed
to seek medical care in the event they experience symptoms (p. 20). Additionally, testing
will only be conducted in areas where there has been no detected presence of mosquito-
borne disease by a county or state agency or mosquito abatement district within a month
of testing (p. 20). The Study Director will consult various resources (Centers for Disease
Control and Prevention, state health department websites, mosquito control districts) to
obtain this information if it is not available on the internet (pp. 20-21). Finally, subjects
will be trained to use an aspirator to remove mosquitoes from their skin before the
mosquitoes bite (p. 21).
General risks of being in the field include risks of exposure to wildlife and other biting
insects, risk of injury due to rough terrain, and physical stress associated with being
outdoors for over 12 hours in a hot and humid environment. These will be mitigated
Page 32 of 118
-------
through study staff providing food and beverages to keep the subjects hydrated and
comfortable, having a pavilion at the test site to provide subjects with shade between
exposure periods, avoiding wildlife, encouraging subjects to perform tick checks and to
remove ticks as soon as they are identified, providing first aid assistance to subjects at the
field during the test day, providing standard first aid items (bandages, antiseptics,
antihistamines), and reminding subjects to inform the study staff at any time if they feel
unwell (p. 21). Study staff will identify the closest hospital to each test site and will be
prepared to accompany a subject who requires medical attention.
The study staff will follow local and state health and safety guidelines to mitigate the
risks of Covid-19. For all in-person encounters, subjects will review a set of screening
questions and have their temperature checked prior to continuing with testing-related
procedures (p. 21).
The risk of unanticipated loss of confidentiality will be mitigated through protecting all
individuals' personal information. Subjects will be assigned a code number, which will
be used to identify them on all study-related data sheets. The report will be kept
confidential, and if any results are published the subjects' identities will remain
confidential. Study records will be maintained in locked cabinets, electronic files will be
kept on a password-protected server, and access to personal information will be limited to
SCJ, the study staff, the recruitment firm, the IRB, and the US EPA (p. 22).
Additional risks to subjects include the psychological risks associated with pregnancy
testing and the risk of contracting mosquito-borne diseases during lab-based testing
procedures. Pregnancy testing will be conducted in private and only a single female
member of the research team will discuss the results with the subject.
Practical steps to minimize most subject risks have been described in the protocol. With
the comments from the EPA addressed, risks to subjects will be effectively minimized
and the remaining risks will have a low probability of occurrence.
4. Benefits: This research offers no benefits to subjects. Depending on the results of the
research, it may benefit society by generating reliable repellency efficacy data that could be
used by the EPA to support the registration of insect repellent products containing the
ingredients tested. Registration of effective repellent products could lead to fewer
mosquito bites and reduced incidents of vector-borne illnesses. The Study Sponsor could
benefit financially from the registration of these products and their eventual marketing
and sale.
5. Risk/Benefit Balance: The protocol describes a procedure for generating scientifically
sound data and includes measures to limit the risks to subjects participating in the study.
With the risk mitigation measures put in place and provided that the EPA's comments are
addressed, the remaining risk to subjects is low and reasonable in light of the potential
benefits of the data to society.
6. Independent Ethics Review: The Western Institutional Review Board (WGC IRB) has
Page 33 of 118
-------
reviewed and approved the protocol, informed consent form, and recruitment materials. On
March 3, 2022, the protocol and materials were conditionally approved provided that
changes to the consent forms were made. On March 8, 2022, a Certificate of Action dated
indicated that WCGIRB had approved the protocol, consent forms, and recruitment script.
The WGC IRB is registered with the Office of Human Research Protections
(IRB00000533) and is accredited by the Association for the Accreditation of Human
Research Protection Programs. Satisfactory documentation of the IRB procedures and
membership is on file with the Agency. Documentation regarding IRB approval of the
protocol, consent and recruitment materials has been provided to the HSRB members with
the background materials for this review.
7. Informed Consent: The protocol calls for having the independent recruitment agency
send the consent form by email to candidates who express interest (p. 4). The consent
form notes that candidates will be scheduled for a training session, at which they will
review the consent documents, have the study staff review the form with them, and have
an opportunity to ask questions before deciding to consent to participate (p. 9).
The recruitment script notes that "If you meet the criteria to participate, please review the
attached Informed Consent Form. If you have any questions on the research or have
questions to be answered before you sign the informed consent form, please contact
[ENTER RECRUITMENT FIRM CONTACT NAME]. We will be sure to have all your
questions addressed before you sign the consent form. If you do not have any questions
and are interested in volunteering please complete the informed consent form."
(Recruitment Script for Boston, p. 3)
In all instances, the consent process informs candidates that their participation is
voluntary and that they can withdraw from participation at any time without forfeiting
benefits or compensation.
The protocol does not specify what the consent meeting/training session will cover,
whether it will be held as a one-to-one or group session, whether there will be
demonstrations of all study-related procedures as part of the consent process, and how
candidates' comprehension of the study's procedures will be gauged before they are
invited to complete the informed consent document.
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. Provision is made for discrete conduct of the pregnancy testing
that is required of female subjects on the day of testing.
Throughout the recruitment and consent processes, and again at the start of each test day,
candidates and subjects will be informed that they are free to decline to participate or to
withdraw at any time for any reason.
The protocol notes that subjects will be compensated for their time spent participating in
Page 34 of 118
-------
the study as follows: $100 for the participating in the training meeting and $35 per hour
(rounded up to the next hour) for participation in a test day. If the test day is postponed
due to weather, subjects who were available for the test day will be compensated $60 for
being available and will be invited to participate on the rescheduled test day. Alternate
subjects who are asked to show up at the test site will be paid $100 for their time (p. 13).
Subjects who participate in the dose determination study will be compensated $100 for
their participation in session, which is expected to last 2 hours (p. 22). The amount of
compensation seems reasonable in light of the inconvenience of being in a field and
subject to mosquitoes and hot, humid conditions for 12+ hours. The amount of
compensation does not seem sufficiently high as to unduly impact a subjects' decision to
participate in the research.
Breaks for subjects between exposures and provision of snacks and drinks have been
incorporated into the study design.
Any expenses for injury or illness incurred as a result of study participation will be paid by
the study sponsor (p. 13).
F. 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 the EPA under the pesticide laws.
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 compares with the requirements
of 40 CFR 26 Subparts K and L and the criteria recommended by the HSRB is appended as
Attachment 1. As noted below and in Attachment 1, revisions are needed in order to generate
research that is likely to meet the standards of 40 CFR 26, Subparts K and L.
With the EPA's comments on the consent form and protocol addressed, the consent materials
and process will meet the requirements of 40 CFR 26.1116 and 26.1117. With the protocol and
all associated materials revised according to recommendations from the EPA and the HSRB and
approved by the WGC IRB, the research will likely meet the applicable requirements of 40 CFR
part 26, Subparts K and L. The materials to be used to recruit subjects (advertisements, scripts)
must be provided to the EPA prior to the initiation of the study in order to comply with the
requirements of 40 CFR 26.1125.
G. Ethics Comments and Recommendations
The EPA's ethics comments are provided below. Minor comments on typographical errors have
not been included here. In addition, the EPA has provided ethics-related comments directly on
the screening scripts to be used during recruitment; these are provided to the HSRB as a separate
file. After all necessary changes have been made, the revised protocol and supporting documents
must be resubmitted for review and approval to the overseeing IRB prior to initiating the research.
Page 35 of 118
-------
1. On page 1, revise the first statement as follows: "This study IS NOT intendedfor
submission to US EPA in support of a registration"
2. Revise the document to be a single protocol outlining how the study will be conducted
from start to finish. A researcher unfamiliar with the study should be able to pick up the
document and understand how to carry out the research. The informed consent, statistical
support for sample size, advertising materials, recruitment scripts, and any other study-
related support materials should be included as appendices to the protocol document. See
the science review recommendations for a suggested organization.
3. Sample size - The EPA recommends a sample size of 13 based on an EPA-conducted
power analysis. The inclusion of an additional subject is unnecessary to ensure an
adequate sample size for statistical analysis and unnecessary exposure of subjects is
unethical.
4. Stopping rule - Include an additional stopping rule indicating that testing will be stopped
when more than half (7 out of 13 treated subjects or 8 out of 14 treated subjects)of treated
subjects experience FCL. At that point, the mCPT can be calculated and further exposure
of human subjects is unnecessary and therefore unethical.
5. Provide more details about the bug suits discussed in the protocol. The EPA recommends
against the use of non-breathable bug suits. Site monitoring, capturing mosquitoes that
land on subjects, training the subjects on aspirator use to catch mosquitoes before they
bite, and providing a screened enclosure for subjects to use between exposure periods
effectively minimize the risk of contracting mosquito-borne illnesses. The use of non-
breathable bug suits presents additional, unnecessary risks of heat exhaustion and heat
stroke.
6. The EPA recommends against having the consent meeting and the subject training
session (attractiveness testing and aspirator use) conducted concurrently or sequentially
on the same day. Separating these into two distinct interactions with the study team
would reduce the pressure for candidates to enroll and would give them adequate time to
consider whether they want to participate after going through the consent process with
the Study Director. It is possible to hold virtual consent sessions and to have individuals
sign the form electronically or in person when they participate in the in-person training
session. No one should not participate in any aspect of the testing prior to giving fully
voluntary, informed consent.
7. Eligibility criteria
a. Employees and family members of any employee of an entity involved in the
study (e.g., recruitment firm, contract research organization) should be excluded
from participation.
b. Clarify whether SCJ will provide unscented soap for subjects to use and whether
subjects may use fragranced products immediately prior to their participation in
the test (p. 3).
Page 36 of 118
-------
Confirm whether subjects will be permitted to participate to test both repellents
and to test at both sites, or whether subjects are limited to testing a single repellent
at a single site.
Although the EPA's guidelines recommend that subjects are between 18 and 55,
feedback from the HSRB and discussions of previous protocols have changed the
recommendation to remove upper age limit or provide rationale for excluding
subjects over 55 years old. Please revise the protocol to address.
Specify the specific health conditions that would make candidates ineligible to
participate and/or how information will be obtained from the potential subjects
and evaluated by the Study Director.
The eligibility criteria limit the participation to English speakers. The research
does not offer benefits to subjects so lack of access to the research does not
deprive non-English speakers of any benefits from participation. The research
may also be generalizable to the US population regardless of the language spoken
by the subjects. However, enrolling non-English speakers would make the study
enrollment more equitable and should be considered by the researchers. Revise
the protocol to describe how subject selection will be equitable given enrollment
is limited to those who speak English.
8. Pregnancy testing - Ensure that throughout the protocol and consent form it is clear that
female subjects will be subject to pregnancy testing prior to any exposure to mosquitoes
or the test substance (p. 20), rather than only on the test days (pp. 4, 8).
9. Recruitment
a. Recruitment should not be initiated until after the protocol has been reviewed by
the EPA and the HSRB, any recommendations have been addressed, and the final
version has been approved by the WCGIRB.
b. Provide more specific details in the protocol on how recruitment will be
conducted using the recruitment firms. How will recruitment firms identify
individuals who might be willing to participate in the study? Will advertisements
be posted to which people can respond? Will the subjects be drawn from a pool of
individuals who have expressed open willingness to be contacted about
participating in the study?
c. Clarify in the protocol how email will be employed in the recruitment process.
What advertisement will be emailed and to whom will it be emailed? How will
the distribution list be established?
d. Include in the protocol measures that will be taken to ensure that subject selection
is equitable.
e. Amend protocol to note the total number of subjects that will be recruited for
participation to ensure that a representative sample (age, gender, race/ethnicity)
can be selected for the field testing on each day. If the protocol calls for 20
subjects per test (14 treated subjects, 2 untreated control subjects, 4 alternate
subjects), consider how many additional candidates should be recruited and
screened to ensure that there is a sufficiently large pool to choose from. It is
reasonable to recruit a pool of potentially eligible candidates that is double the
c.
d.
e.
f.
Page 37 of 118
-------
total number of subjects needed for that test day/site, e.g., a pool of 40 candidates
to fill the 20 slots needed for testing the PMD product in Wisconsin.
f. All recruitment materials, including advertisements, must be provided to EPA and
approved by the IRB prior to use.
10. Informed consent document and process
a. Remove the consent form from the body of the protocol and include as an
attachment to the protocol.
b. Expand the discussion of the consent process in the protocol. It should include
more specific details about what will be covered (the entirety of the consent form
rather than the inclusion/exclusion criteria), as well as whether it will be a one-on-
one or group session, whether individuals will be able to ask questions of the
Study Director in private, and where it will be conducted.
c. The consent process should include a demonstration of all of the procedures that
will be used in the study, including attractiveness testing, aspirator use training,
dosimetry testing, product application, and an exposure period on the test day.
d. Individuals should be given time to consider whether they want to participate
before being required to provide consent.
e. Before inviting subjects to sign the consent form, the Study Director or person
conducting the consent meeting should ensure that the subject has understood the
materials provided in writing, presented, and demonstrated. Include the method
by which potential subjects' comprehension will be evaluated in the protocol.
f. The consent form must be revised to include all relevant elements of consent
required under 40 CFR 26, Subpart K, including the simple summary of the study
and the identity of the test substances.
g. The consent form and discussion must identify the test substance(s). Information
such as the SDS should be available to subjects.
h. Delete the statement that "If you have a serious reaction to the product, you will
be told what ingredients are in it." (p. 12) All subjects will be informed of the
active ingredients in the test products. Any subject who experiences any reaction
to the test substance may need be informed of all of the products' ingredients if
necessary to provide medical treatment.
i. Add a statement that evaluating adverse effects may require the study personnel to
consult with the treating medical personnel, after obtaining the subject's consent.
For example: "The Sponsor, medical monitor, and the Study Director will
determine whether the injury is related to the subject's participation in this study.
To do this, they may request your consent to consult with the person/facility that
provided medical treatment following an adverse effect, which could require your
consent."
j. The consent form should be revised to align with all changes made to the protocol
and to address all of the EPA's specific comments on the consent form.
k. Clarify what "five individual tests" will be carried out during the study, and
whether subjects can participate in more than one day.
1. Revise the consent form to reflect the number of subjects in the study.
m. Indicate whether the data from subjects who withdraw will be included in the
study results, and if so, under what conditions.
Page 38 of 118
-------
All subjects should receive a copy of signed consent form (printed or electronic),
without requesting it from the recruitment firm (p. 14).
The consent form should include one of the following statements about any
research that involves the collection of identifiable private information or
identifiable biospecimens:
(i) A statement that identifiers might be removed from the identifiable
private information or identifiable biospecimens and that, after such
removal, the information or biospecimens could be used for future
research studies or distributed to another investigator for future research
studies without additional informed consent from the subject, if this might
be a possibility; or
(ii) A statement that the subject's information or biospecimens collected as
part of the research, even if identifiers are removed, will not be used or
distributed for future research studies.
11. Attractiveness testing and training on aspirator use
a. Regarding mosquitoes used in the attractiveness testing and aspirator training,
please clarify whether the mosquitoes will be destroyed after use in each test or
whether mosquitoes will be reused in either the attractiveness testing or aspirator
use training.
b. Specify in the protocol how the disease-free status of mosquitoes used in lab-
based testing will be confirmed/documented.
c. Revise the protocol to indicate whether subjects who make more than one attempt
during the attractiveness testing use the same batch of mosquitoes or a new batch.
12. Risks to subjects - This section should be revised to appropriately identify all risks and to
indicate how known risks to subjects will be minimized.
a. Risks of exposure to the test substance will be minimized by excluding subjects
with skin conditions that could be exacerbated by exposure to the test substance
and subjects with known allergies to insect repellent products.
b. Exposure to biting mosquitoes should be minimized by having subjects wear
appropriate gear, such as hat with head net, gloves, long-sleeved shirt, pants, and
closed-toed shoes or boots. As noted in the science review and the
recommendation sections about the eligibility criteria, the EPA recommends
against the use of non-breathable bug suits, which are similar to Tyvek suits, in
field testing due to the significant risks of heat-related illness when worn for long
periods in hot, humid environments. Further, this risk can be mitigated by
conducting all non-field testing study procedures (training, attractiveness testing,
dosimetry) in a lab setting. Provide more information about the bug suit proposed
to be used during testing and measures that will be taken to reduce any additional
risks of heat stress/heat-related illness if subjects wear a bug suit during testing.
c. In addition to coordination with the state and local public health districts and
mosquito abatement districts, risks to subjects should be minimized through
trapping at the test site for at least one month prior to the field testing, and testing
collected mosquitoes to identify any mosquito-borne pathogens that can be
transmitted to humans.
n.
o.
Page 39 of 118
-------
d. The pavilion provided at the test site should be fully screened to keep mosquitoes
and other insects out. Seating should be provided for subjects to rest between
exposure periods to minimize the risk of exhaustion.
e. Exposure to biting mosquitoes in the lab during aspirator training and testing
should be mitigated through the use of batches mosquitoes from a lab-reared
colony that have been certified as pathogen-free. The pathogen-free certification
should indicate the specific diseases for which the mosquitoes in the batch/colony
were screened and when the screening occurred.
f. Consider including the risks of heat-related illness and how they will be mitigated
in the protocol, and including in the consent form and/or in the discussion at the
beginning of each test day information for subjects about how to identify
symptoms of heat-related illnesses.
g. Specify who will provide "qualified first aid assistance" to the subjects during the
field testing and what the qualifications are.
13. Include a specific section in the protocol on how adverse events will be identified,
evaluated, and reported to the IRB. This should explain who will be determine whether
adverse events are related to participation in the study and whether they are serious, and
the qualifications of the person or team making the determinations.
14. The protocol should include language about the Sponsor's agreement to pay for any
medical expenses associated with treating a study-related illness or injury.
15. The protocol should indicate whether the study will have a medical monitor, and the
specific role the medical monitor will play during each test day. Additionally, if any
medical personnel (e.g., doctor, nurse, emergency medical technician, certified first aider)
will be present for each test day, specific the person, their qualifications, and role at the
test site.
16. Compensation
a. Indicate whether individuals will be compensated for participating in a consent
meeting regardless of whether they enroll, and if so, how much.
b. Specify whether subjects when subjects will be compensated, e.g., at the end of
their participation or at the end of each interaction with the study team.
c. The recruitment forms specify a difference in compensation for alternate subjects;
one indicates that subjects would receive $50 for being an alternate and the other
indicates that subjects would receive $100 for being an alternate. Revise to be
consistent with the protocol, which indicates all alternates who are not enrolled as
test subjects will receive $100.
17. Dosimetry
a. The protocol calls for establishing a typical consumer dose for the picaridin stick
product through dosimetry testing in a laboratory or at a field site prior to
conducting any field efficacy testing is performed. Dosimetry is ethically justified
for the picaridin product because no standard dose for field testing has been
Page 40 of 118
-------
established, and such a dose can be established through the observation of human
subjects making typical applications to their own skin.
b. The protocol also notes that subjects would not be allowed to treat themselves
with more than 2.0 g product/600 cm2 during dose determination testing. Limiting
the dose applied during the dose determination phase is impractical and unethical,
because it would not accurately reflect the amount of product typically applied by
consumers using the product and would bias the resulting dose calculated for use
in field testing.
c. If dosimetry testing for the picaridin product is pursued, the protocol should be
revised to indicate a statistically-supported sample size for conducting a
dosimetry test. In the past, EPA has recommended a sample size of 25. If data are
not scientifically valid, then it is not ethical to rely on the data. Additionally, SCJ
should consult with EPA prior to using the dose derived through dosimetry to
confirm that it is an appropriate level for field testing.
18. Field site selection and monitoring
a. Prior to field testing, consistent mosquito landing pressure for the proposed
duration of testing should be confirmed in order to avoid engaging subjects in
testing and applying repellents unnecessarily. The protocol calls for basing the
site selection weekly testing conducted by SCJ and the Norfolk County Mosquito
Control District (p. 17); this should be revised in accordance with the EPA's
science comments. Trapping before testing occurs ensures that there are diverse
mosquito species and screening for diseases prevents testing where a known
vector-borne illness has been identified. Consistent adequate landing pressure
allows a test day to proceed to completion, thereby minimizing human exposure
to both the test substance and mosquitoes to the greatest extent possible.
b. Revise the protocol to include screening the mosquitoes captured at the test site
for pathogens that could be carried by the species present in order to increase the
likelihood of identifying any mosquito-borne diseases present at the test site and
to further minimize the risk of subjects being exposed to mosquito-borne diseases
during the field testing.
c. Identify all potential test locations in the protocol and explain how site monitoring
will be conducted at each of the sites. The addition of any site not listed in the
protocol will require an amendment to the protocol submitted to and approved by
the IRB before any work at the site can begin. The amendment should provide in
detail the information about how the site will be monitored prior to the field
testing day(s), the process for trapping and screening mosquitoes, and the
appropriate agencies that will be consulted for information about vector-borne
illness in and around the additional test location.
19. Test day preparation
a. Provide more detail about the preparation for and timing of the study day. Include
details such as whether subjects will be reminded of the restrictions before and
during the testing (and if so, how and when) and how far in advance of the field
testing start time and/or at what time will subjects arrive at the test site to receive
treatment.
Page 41 of 118
-------
b. During the initial briefing and test day compliance check, research staff should
remind subjects that they are free to withdraw at any time and that they will be
compensated for their participation up to the time of their withdrawal.
20. Test days
a. Subjects must be assigned randomly as control or alternate subjects. Revise the
protocol to delete the provision allowing the Study Director to choose untreated
control subjects based on their proficiency using an aspirator.
b. Subjects' skin should be examined for disqualifying conditions prior to
application of the test substance and again at the end of the test period.
c. If participants are eligible to participate in more than one test day, space the two
test days a minimum of 72 hours apart to minimize discomfort to subjects.
d. Mosquito landing pressure measured by the untreated control subjects must be
established as 5 landings within 5 minutes on each of the control subjects in order
for EPA to consider the data reliable. It is not ethical to rely on data that are not
scientifically acceptable; ensure that the protocol language for landing pressure
matches the definition provided by EPA in the science review comments.
e. If the test is moved "to a nearby location where biting pressure may be higher",
the subjects should still have the same access to a screened enclosure for resting
between exposure periods.
f. Subjects should be reminded at the start of the test day that they are free to
withdraw from the study at any time and without penalty, and whether/how their
data would be used if they choose to withdraw after testing has begun.
g. The EPA recommends instructing all subjects to wash their treated limb as soon
as possible after their participation ends and providing soap, water, and paper
towels at the field site or at the facility where subjects will be returning to at the
end of the testing day.
21. To confirm that subjects did not contact a vector-borne illness or suffer adverse events
after the field testing, the protocol should be revised to include a provision that that the
Study Director or staff will contact all subjects after their participation. This contact
should be made after enough time has elapsed that symptoms of vector-borne illness
would have appeared. If a subject reports an adverse effect that is likely related to
exposure to a vector-borne illness and their participation in the test, other subjects who
participated in the same test day should be notified.
22. The protocol calls for notifying the IRB within 10 days if a mosquito-borne disease is
detected in the mosquitoes collected on test days (p. 17). The approval from the WCG
IRB notes that "any site monitoring report that directly and materially affects subject
safety or their willingness to continue participation" should be provided to the IRB within
5 days (WCG IRB Certificate of Action, 3/8/2022, p. 2). The protocol should also include
a timeframe for notifying subjects in the instance any mosquito-borne illness is detected
at a test site where they were present.
23. Protocol amendments and IRB oversight
Page 42 of 118
-------
a. Revise the protocol discussion on amendments to the protocol (p. 5). Any changes
to the protocol must be submitted to the IRB for review and approval before
implementation. Amendments should not be signed and dated as effective by the
Study Director until after receiving approval from the IRB.
b. A protocol change intended to eliminate an apparent immediate hazard to subjects
may be implemented immediately provided WGC IRB is notified according to its
procedures.
c. All amendments, deviations, and any adverse events must be documented in the
study and reported to the WGC IRB consistent with its reporting procedures.
d. Any changes to the protocol should be reflected in the protocol itself, rather than
as a note to file or attachment.
e. Ensure all references to the IRB overseeing the research are accurate in the
protocol and consent materials. The consent form notes that Schulman Associates
IRB may be contacted (p. 14).
f. Include a statement that "This study will be conducted in accordance with the
EPA's final regulation at 40 CFR 26 that establishes requirement for the
protection of subjects in human research. The protocol, informed consent
form, and other required documentation for this study must be approved by
an independent institutional review board and submitted to the EPA as
required by 40 CFR 26.1125 before the research can be initiated. The report
of the completed research is subject to the requirements at 40 CFR 26.1303
to provide documentation related the ethical conduct of the study."
24. COVID-19 - Include that all federal regulations/guidelines in effect will be followed, as
well as a process to notify study staff and/or subjects if anyone with whom they had
contact during the study becomes ill.
25. Confidentiality - If photos will be taken at any point during the study, revise the protocol
to include a statement that any photos taken during the study will not include the
subjects' faces or identifying features, or that these elements will be blurred before the
photos are distributed or published.
26. Delete or provide a rationale for allowing the Study Director or other delegated staff to
"end a particular participant's participation on a test day at any time, for any reason"
(Protocol, p. 22). Generally, the EPA recommends that the Study Director have limited
discretion to withdraw subjects from the study as outlined in the protocol. For example,
such discretion could be characterized as: "Participants' enrollment in the study may be
ended at the discretion of the Study Director where continued participation may affect
the safety of the participant or where there is a development of any condition that might
interfere with study participation
27. Documentation
a. The protocol should include sample data collection sheets and other forms that
will be used to collect and track subject information.
Page 43 of 118
-------
b. EPA recommends that the researchers consider the guidance from the Food and
Drug Administration when drafting the final report.15 Specifically, pages 7-9
provide a summary of how a study report should be organized and what it should
contain to be free from ambiguity and to facilitate review.
Attachments:
1. EPA Protocol Review
2. Ethics Review Checklists
3. SCJ Statistical Support for Sample Size
4. EPA Power Analysis
5. Basis for EPA Recommended Standard Dose
Page 44 of 118
-------
ATTACHMENT 1
EPA Protocol Review
Title: Science and Ethics Review of a Protocol for Efficacy Testing of SCJ Personal
Repellent Products against Mosquitoes in a Field Study.
Date on Draft Protocol: February 2022
Principal Investigator: Daniel Usry, S.C. Johnson & Son Institute of Insect Science for
Family Health
Sponsor:
S.C. Johnson & Son, Inc.
1525 Howe Street
Racine, WI 53403
Sponsor Contact:
Ed Whittle
S. C. Johnson & Son, Inc.
1525 Howe Street
Racine, Wisconsin 53403
Phone: +44 1276 852832
Recruiting Agency:
Reckner (Wisconsin site)
9833 South 13th Street
Oak Creek, WI
Phone: 414-768-6040
Focus on Boston
30 Rowes Wharf
Boston, MA 02110
Phone: 617-737-0227
Testing Locations:
Brighton Dale Links Golf Club (Wisconsin Site)
830 248th Avenue
Kansasville, WI 53139
Contact: Shelley Carlson - 414-768-6057
Hale Reservation (Massachusetts Site)
80 Carby Street
Westwood, MA 02090
Contact: Larry Jenkins - 781-356-7318
IRB:
Page 45 of 118
-------
WCGIRB
1019 39th Ave. SE, Suite 120
Puyallup, WA 98374
Telephone: 855-818-2289
E-mail: researchquestions@wcgirb. com
1. Societal Value of Proposed Research
(a) What is the stated purpose of the proposed research?
The objective of this study is to determine the duration of efficacy of two different
personal repellents against mosquitoes that have the potential to transmit
pathogens of public health importance in a field setting using human subjects.
Efficacy will be measured as the duration between test substance application and
the first confirmed mosquito landing on a treated subject's treated appendage. The
FCL is defined as the time at which one mosquito landing occurs and a second
landing occurs within 30 minutes. The second landing confirms the first landing.
The period from treatment to the FCL will be defined as the CPT. An mCPT will
be calculated from the CPTs of the treated subjects for each product.
(b) What research question does it address? Why is this question important?
Would the research fill an important gap in understanding?
CPT data points will be statistically analyzed to determine mCPT for each product
tested. This information does not currently exist for the proposed product
formulations containing PMD and picaridin as their active ingredient. The data
generated by the proposed research will be used to characterize the proposed
products' repellency duration (mCPT) in support of product registration with the
EPA. The proposed study will characterize the efficacy of the products for
registration, which has not yet been evaluated.
(c) How would the study be used by EPA?
The proposed study would be used to generate product-specific efficacy data
which the EPA requires for registration of skin-applied insect repellents. As of this
writing, the product formulations containing PMD and picaridin as an aerosol or
solid stick have not yet been evaluated by the EPA regarding their performance
against mosquitoes of public health importance. The data that is generated will be
reviewed to assure that all requirements are met regarding use of these skin-
applied insect repellents by humans and that any claims made on their label are
supported.
A standardized protocol evaluated and approved by the Agency will ensure that the
data is generated in accordance with the appropriate EPA Guideline and Rules,
namely OPPTS 810.3700: Insect Repellents to be Applied to Human Skin, the
Repellency Awareness Guidance For Skin-Applied Insect Repellent Producers, and
Page 46 of 118
-------
the most recent Product Performance Rule which can be found at:
https://www.federalregister.gov/documents/2022/04/15/2022-Q7963/pesticide-
product-performance-data-requirements-for-products-claiming-efficacv-against-
certain#sectno-reference-158.1756.
(d) Could the research question be answered with existing data? If so, how? If
not, why not?
The EPA requires product-specific efficacy data to support any product
registration. No previous data has been submitted or approved regarding the use of
these products against mosquitoes under the proposed use pattern.
(e) Could the question be answered without newly exposing human subjects? If
so, then how? If not, why not?
Humans are the target for the test products and there are currently no recognized
reliable models or surrogates available for repellency testing that accurately mimic
the intended use of these products on humans against mosquitoes of public health
concern.
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 two novel insect
repellents, an aerosol containing 10% PMD and a solid stick containing 10%
picaridin as the active ingredients. No hypothesis was explicitly stated in the
protocol.
(b) Can the study as proposed achieve that objective or test this hypothesis?
The objectives may be achieved after revisions to the protocol suggested by the
Agency pertaining to scientific and ethical concerns are implemented. The protocol
does not likely meet the scientific and ethical standards necessary for a successful
product registration with the Agency in its current, unedited form.
2.1 Statistical Design
(a) What is the rationale for the choice of sample size?
Based on a power analysis, the EPA typically recommends a sample size of 13
treated subjects when performing skin-applied repellency studies (Attachment 4),
which has been simulated to provide an adequate amount of data for statistically
meaningful results (i.e., appropriate statistical power). A sample size of 14 treated
subjects is proposed for each field test site. The rationale given by the sponsor
Page 47 of 118
-------
(Attachment 3) is that: [1] a sample size of 14 allows for an equal number of male
and female treated subjects and, [2] 14 test subjects gives a more precise low-end
estimate of the median. After reviewing the results of the simulation of power vs.
sample size, EPA found no notable difference in the power (which was calculated
based on the ratio of the lower bound of 95%CI / estimated median CPT) between
the sample sizes of 13 or 14 subjects (Attachment 4).
(b) What negative and positive controls are proposed? Are proposed controls
appropriate for the study design and statistical analysis plan?
Two untreated subjects, ideally a male and a female, will serve as controls for the
duration of the study at each test location to monitor adequate landing pressure at
the test site during efficacy testing by exposing their untreated limb for 5-minute
exposures. The proposed controls are appropriate for the study design. EPA
recommends that landing data collected from untreated control subjects be solely
used to assess landing pressure during efficacy testing, and that untreated control
subject data should not be statistically compared to data provided by the treated
subjects.
(c) How is the study blinded?
The subjects will not be blinded as to the identity of the test substance that they
are treated. Study staff and subjects will know the identity of the test substances.
(d) What is the plan for allocating individuals to treatment or control groups?
There will be a total of 20 subjects present per test site, including 14 treated
subjects, 2 untreated control subjects, and 4 alternate subjects (without further detail
in the protocol). The Study Director or Principal Investigator will randomly select
an equal number of male and female treated and alternate subjects based on
assigned coded numbers. Untreated control subjects will be selected by the Study
Director from those that are proficient at aspirating landing mosquitoes. The
randomization procedure for subject assignments is not described in sufficient
detail. EPA's recommendations for randomization procedures are included in
comment # 12 of the science review.
(e) Can the data be statistically analyzed?
Yes. See (f) and (g) below.
(f) What is the plan for statistical analysis of the data?
It was proposed that''for each formula tested, the median CPT across all the test
subjects will be calculated as the summary measure of CPT." EPA recommends
adhering to the procedures outlined in the Repellency Awareness Guidance For
Skin-Applied Insect Repellent Producers
Page 48 of 118
-------
(https://www.regulations.gov/document/EPA-HQ-OPP-2013-0406-00Q3) for
determining the CPT used for product labeling purposes.
(g) Are proposed statistical methods appropriate to answer the research
question?
It was proposed that''for each formula tested, the median CPT across all the test
subjects will be calculated as the summary measure of CPT." Due to imprecise
statistical language in the protocol, the appropriate statistical methods are
recommended by EPA in comment #13 of the science review.
(h) Does the proposed design have adequate statistical power to definitively
answer the research question?
Yes, the proposed sample size of 14 subjects has adequate statistical power based
on the results of an EPA simulation/power analysis (Attachment 4) and the
purported registrant simulation (Attachment 3).
2.2 How and to what will human subjects be exposed?
During the field efficacy studies, treated subjects will be exposed to a 10%
picaridin solid stick or a 10% PMD aerosol formulation. It is currently unclear if
the same treated subjects will have the ability to participate in more than one test
on a different day. The aerosol product containing 10% PMD will be applied at a
dose of 1.0 g product/600 cm2 of skin to either a treated subject's arm or leg.
Dosimetry testing in a laboratory or at a field site will be performed for the solid
stick picaridin formulation before any field efficacy testing is performed and the
protocol states that "//o test subjects would be allowed to treat themselves with a
greater than 2.0 g 600 cm2 application rate during dose determination testing.
Field exposure time was reported as potentially 12+ hours for each day of testing
per subject, but the Agency recommends a distinct cutoff number of hours. In the
field, subjects will be exposed to natural mosquito populations relevant to public
health where pathogens have not been detected for at least one month prior to
testing. In the laboratory during pre-test attraction testing and aspirator training,
subjects will be exposed to lab-reared mosquitoes that have not taken a blood
meal. The EPA recommends that molecular-based testing is conducted to certify
that all mosquitoes used in laboratory procedures are pathogen free.
(a) What is the rationale for the choice of test material and formulation?
Efficacy data are required to characterize product performance and support
efficacy claims against mosquitoes on product labels. EPA requires submission of
product performance data for registration of all products claiming efficacy against
public health pests.
Page 49 of 118
-------
(b) What is the rationale for the choice of dose/exposure levels and the staging of
dose administration?
Based on an analysis of dosimetry results from repellent studies reviewed by EPA
and HSRB since 2006 (Attachment 5), EPA considers a dose of 1.0 g product/600
cm2 of skin to be an appropriate product dose for testing aerosols, wipes, and
lotion type products and 0.5 g (± 10%) product/600 cm2 of skin for testing pump
spray type products. The EPA recommended that this rationale be include in the
protocol. EPA does not have a recommended dose for use with the solid stick
product based on existing dosimetry data, so dose determination testing to
establish a typical consumer dose will take place for the picaridin product before
field efficacy studies begin.
(c) What duration of exposure is proposed?
Dosimetry will only be performed for the picaridin solid stick. In the laboratory
(for dosimetry), subjects will be instructed to apply the test product to their own
arm or leg based on proposed labeling directions. This dose determination testing
will occur presumably over a matter of minutes, but the duration was not explicit
in the proposed protocol. Exposure time in this instance will be comprised of the
time it takes to apply the product to achieve complete coverage and then washing
the product off once sufficient application is completed.
Field testing was described as potentially taking 12+ hours per day, but no distinct
cutoff time was established. Repellency testing for each product will take one day
per field site, but it is unclear if both products will be tested on the same day at
the same site or if the same subjects will be used between different test days.
Exposure time will begin when the product is sprayed (PMD aerosol) or wiped
onto the treated subject (picaridin stick) and will end when the FCL occurs, until
testing ends after "12+ hours," or the end of testing is prompted by a stopping
rule. The Agency recommended that a discrete maximum testing duration be
clearly defined in the protocol.
Proposed exposure periods consist of exposing treated skin to field mosquitoes for
5 minutes at 30-minute intervals until CPT is reached by the treated subjects or
end of testing, whichever occurs first. Untreated control subjects will monitor
landing pressure at the test site by exposing an untreated limb to wild mosquitoes
for 5 minutes, every 30 minutes. When untreated control subjects have received
five landings in a five-minute exposure period, they will cover their exposed
limbs by rolling down their sleeve or pant leg. It is unclear if each five-minute
exposure of untreated control subjects will precede each five-minute exposure of
treated subjects.
The Agency recommended revising the protocol language to specify that each 30-
minute interval should start with untreated control subjects exposing their
untreated limb (for five landings or five minutes, whichever occurs first),
Page 50 of 118
-------
immediately followed by treated subjects exposing their treated limb for < five
minutes (until a landing is observed or for five minutes, whichever occurs first
(see Recommendation #9, part (d)).
2.3 Endpoints and Measures
(a) What endpoints will be measured? Are they appropriate to the question(s)
being asked?
Efficacy will be measured as the duration between the time the test substance is
applied and the time the first confirmed mosquito landing on a treated subject's
treated appendage occurs which will be defined as CPT. FCL, which is the
endpoint of the study for each treated subject, is defined as the time at which one
mosquito landing occurs and a second landing occurs within 30 minutes. The
second landing confirms the first landing. The period from treatment to the first
confirmed landing will be defined as the CPT. The endpoints are appropriate for
the question being asked.
(b) What steps are proposed to ensure measurements are accurate and
reliable?
• Compliance with Good Laboratory Practices (GLP) as defined by 40 CFR 26,
Subparts K-L.
• Specific inclusion and exclusion criteria will be established and followed
closely so that testing will be performed on a representative population of
insect repellent users in the United States.
• EPA's standard dose of 1.0 g product/600 cm2 of skin will be used for the
PMD aerosol testing; a dose determination test (dosimetry) will be conducted
in the laboratory for the picaridin stick product.
• Limb measurements will be conducted to calculate the appropriate treatment
area for each subject.
• Subjects will be identified by numbers that will be used to randomly select for
the various groups and products that are being tested. However, the
randomization details need to be discussed in more detail in a revised protocol.
• Attractiveness testing and aspirator training are proposed to be conducted in
the field, but the Agency recommends performing these procedures only in the
laboratory.
• Lab or field procedures were proposed to assess a subject's attractiveness to
lab-reared mosquitoes. Subjects that are deemed unattractive (not receiving an
adequate number of landings within a given time frame) will not participate in
further training or testing. The Agency recommended that these procedures be
solely performed in the lab, not in the field.
• Lab or field procedures were proposed to instruct subjects on how to properly
aspirate lab-reared mosquitoes after they have landed on their skin, but before
they are able to bite. However, further details will need to be provided in a
revised protocol that describe the specific criteria at which a subject is
Page 51 of 118
-------
sufficiently proficient to participate in field testing. The Agency recommended
that these procedures be solely performed in the lab, not in the field.
• Lab-reared mosquitoes used for attractiveness testing and aspirator training
will be certified pathogen-free and readily available for the number of
replicates deemed necessary to assess attractiveness and achieve sufficient
proficiency.
• Fourteen treated subjects, half male and half female, will be treated with the
product (either the PMD aerosol or picaridin stick) and exposed to wild
mosquitoes until the FCL for the product occurs.
• Two untreated control subjects, preferably one male and one female, will serve
to monitor adequate mosquito landing pressure at each test site.
• Four alternate subjects will be available for each day of testing in case a treated
subject has to withdraw from the test, or subjects need to be replaced for any
reason. If no one chose to withdraw from the test, alternates will be excused
from the test site once the last treated subject has received their repellent
applications.
• Efficacy testing will be conducted at two geographically separate sites.
Mosquito surveillance and pathogen testing will occur to ensure that pathogens
are not detected at field sites at least one month prior to efficacy testing and to
ensure that the required species for testing are likely to be present at the
selected field sites during testing.
• Mosquitoes that land on any of the " test subjects" on the exposed limb over the
duration of the field study will be collected for taxonomical identification and
pathogen screening. However, it is unclear in the protocol if the phrase " test
subjects" refers to treated subjects only or all subjects in this context.
• A Quality Assurance Unit will inspect"at least one phase of the testing to
ensure the integrity of the study
• A final report will be reviewed against the protocol, SOPs, and raw data for
accuracy.
(c) What QA methods are proposed?
The Quality Assurance Unit will inspect "at least one phase to ensure the
integrity of the studyIn addition, the final report will be reviewed against the
protocol, SOPs, and raw data for accuracy. A statement will be included in the
report specifying inspection dates, phases, and the dates findings were reported to
the Study Director and Management.
(d) How will uncertainty be addressed? Will point estimates be accompanied by
measures of uncertainty?
Sources of variation include mosquito species and activity at two geographically
distinct sites, and attractiveness of subjects to mosquitoes. These uncertainties
will be addressed by control subjects monitoring landing pressure throughout the
test, by determination of a subject's attractiveness to mosquitoes prior to efficacy
Page 52 of 118
-------
testing, and by using the lowest (most conservative) mCPT per product per field
site for the duration of efficacy evaluation.
3. Subject Selection
3.1 Representativeness of Sample
(a) What is the population of concern?
The population of concern is any person that are users of skin-applied repellent
products to protect themselves from mosquito bites in the United States.
(b) From what populations will subjects be recruited?
Candidates will be recruited from a population of healthy males or females, ages
18-55, in the United States. The protocol also stipulated that candidates provide
proof of age by a driver's license, passport, or other valid form of identification
and must be able to read and understand English. The protocol indicates that an
equal number of male and female subjects is desired.
(c) Are expected participants representative of the population of concern? If
not, why not?
Yes. Recruitment is intended to capture a pool of candidates that reflect the
population of skin-applied repellent users in the United States.
(d) Can the findings from the proposed study be generalized beyond the study
sample?
Yes, if the protocol is revised to include two field testing sites that are
geographically distinct, where the predominant mosquito species differ.
Furthermore, the taxa across the final test sites used must include all
representative mosquito species required for testing under the Product
Performance Rule to support mosquito repellency claims. The study should also
include a statistically adequate number of replicate subjects for each test site.
Power analysis is employed for estimation of adequate sample size to calculate a
reliable estimate of mCPT, and the randomization plan requires maintaining an
even distribution of male and female subjects to each treatment group.
3.2 Equitable Selection of Subjects
(a) What are the inclusion/exclusion criteria? Are they complete and
appropriate?
The eligibility criteria are included on pages 3-4 of the protocol. They include:
Page 53 of 118
-------
Inclusion Criteria
1. Must be between ages 18-55 and provide proof of age by a driver's license,
passport, or other valid identification
2. Must read and understand English
3. Must not be employees of SCJ or immediate family members of SCJ employees
4. Must have a reliable form of transportation to get to andfrom the test and
training locations. Subjects are responsible for their own transportation to and
from the training and test locations
5. Must be willing to be exposed to and potentially bitten by mosquitoes. Must not
be known to be hypersensitive to mosquito bites. All measures possible will be
taken to prevent mosquito bites.
6. Must be willing to refrain from using alcohol 12 hours before the test, and
refrain from nicotine, andfragrance products (e.g., soap, perfume, cologne, hair
spray, lotion, etc.) during the test
7. Must be willing to follow the study procedures as explained and be willing to
sign an ICD
8. Due to Covid-19, Subjects must be willing to have their temperature checked
via a contactless infrared digital thermometer prior to participating each training
or test day(s).
9. Must be a user of insect repellent products
Exclusion Criteria
1. Sensitivity or allergy to mosquito bites, Elastikon (or equivalent) tape, latex
insect repellents, or skin care products
2. Suffer from respiratory problems such as asthma
3. Currently suffering skin disease or skin problems, such as eczema, psoriasis, or
atopic dermatitis
4. Health conditions or any health concerns that would make them unable to
remain outdoors, in a suit made of tightly woven material that a mosquito's
proboscis cannot penetrate (a bug suit), for several hours where high
temperatures, high humidity and sweating are likely to occur
5. Health conditions or any health concerns that would make them unable to sit in
a chair for long periods, with breaks for limb stretching and movement at
reasonable intervals, or unable to stand continuously for five minutes in
conditions where high temperatures, high humidity and sweating are likely
6. Female subjects must not be pregnant or breast-feeding. Female test subjects
will be required to take a pregnancy test whenever they will be exposed to the test
substance or mosquitoes. Female test subjects will be required to arrive at the test
location at an earlier time than the male subjects. At this time a female study staff
member will discuss with each female subject whether a pregnancy test will be
needed Any female subject that affirms they are naturally or surgically incapable
ofpregnancy will not be required to take a pregnancy test. For those female
subjects that could potentially be pregnant, it will be required to perform an over-
the-counter pregnancy test the day of the test. The pregnancy test will be supplied
by the Sponsor. The test will be performed by each test subject alone in a private
Page 54 of 118
-------
bathroom. After completion of the pregnancy test, a female member of the study
staff will ask each female test subject separately in a private setting, away from
all other people, if the potential subject is still interested in participating in the
study. If the test is negative and the test subject is interested in participating, the
results will be verified by a female of the study staff in a private manner.
These criteria are also included on pp. 19-20 of the protocol. EPA has
recommended the addition and clarification of some eligibility criteria.
The eligibility criteria limit the participation of subjects to English speakers.
Though the research does not offer benefits to subjects, enrolling non-English
speakers would make the study enrollment more equitable.
(b) What, if any, is the relationship between the investigator and the subjects?
There should be no relationship between the investigator and subjects beyond
their interactions as part of the research study. Those who are employees of SCJ
or immediate family members of SCJ employees are prohibited from participating
in the study (p. 3).
(c) Will subjects be recruited from a vulnerable population?
The protocol states "//o one categorized as 'vulnerable' will be considered in this
placement' (p. 4).
Recruitment will be conducted in two different locations within the United States
and will be conducted by independent recruitment firms. The recruitment target is
to assemble a pool of potentially eligible subjects who represent the demographics
of users of skin-applied mosquito repellents. The EPA does not believe that based
on the protocol and recruitment plan, recruitment will target subjects from
vulnerable populations.
(d) What process is proposed for recruiting and informing potential subjects?
The protocol notes that "subjects will be recruitedfrom the general public by an
independent recruitment agency with the goal that the pool closely represents the
general demographics of repellent users in North America, aiming for an equal
number of male andfemales" (p. 4). "The recruitment firms will contact subjects
via internet platforms following the screening document approved by the IRB.
Potential subjects will be given a brief outline of the study, their role in the study,
and the inclusion exclusion [sic] criteria. If they meet the criteria and are
interested in potentially enrolling in the study, they will notify the recruitment
firm and they will be scheduled to attend a test subject training session lead by
the SCJ study staff' (pp. 18-19).
Page 55 of 118
-------
The protocol is unclear how exactly candidates will be recruited, only noting that
they will be contacted by email by the recruitment firm. No advertisements or
other recruitment materials beyond the email template were provided. EPA has
made recommendations regarding the recruitment process and the manner in
which subjects are informed about the study that should be addressed before the
research is initiated.
(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 by independent recruitment agencies, minimizing the
potential for coercion or undue influence related to the decision about whether to
participate in the study. In addition, employees of the study sponsor and
employees' immediate family members are excluded from participation,
safeguarding this group from undue influence to participate in the study.
Employees of the research firms and any other organization associated with
conducting this research study, as well as the employees' family members, should
also be excluded from participation in the study in order to avoid any appearance
of coercion or undue influence to enroll.
3.3 Remuneration of Subjects
(a) What remuneration, if any, is proposed for the subjects?
The protocol notes that subjects will be compensated for their time spent
participating in the study as follows: $100 for the participating in the training
meeting and $35 per hour (rounded up to the next hour) for participation in a test
day. If the test day is postponed due to weather, subjects who were available for
the test day will be compensated $60 for being available and will be invited to
participate on the rescheduled test day. Alternates who are asked to show up at the
test site will be paid $100 for their time (p. 13). Subjects who participate in the
dose determination study will be compensated $100 for their participation in
session, which is expected to last 2 hours (p. 22).
(b) Is proposed remuneration so high as to be an undue inducement?
No, the remuneration is reasonable in light of the burden and inconvenience
associated with participated in the study.
(c) Is proposed remuneration so low that it will only be attractive to
economically disadvantaged subjects?
No. The rates seem reasonable for the burden associated with spending 12 plus
hours in a hot, humid environment and subjecting oneself to mosquitoes.
Page 56 of 118
-------
(d) How and when would subjects be paid?
Subjects will receive their compensation by check or prepaid gift card from the
independent recruitment agencies. The protocol is unclear when compensation
will be provided - either at the end of every subject interaction with the study
team, or at the conclusion of their participation. The EPA has requested that this
information be included in the protocol.
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?
One of the test materials is a spray product that contains 10% PMD. A 90-Day
dermal study in rats (MRID 44438710 40) 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 44438711) 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.
The other test material is a solid stick product containing 10% picaridin. The
active ingredient picaridin is classified as toxicity category IV for Acute Oral
(LDso >5,000 mg/kg (MRID 51868905 and 44408748), Acute Dermal (LDso
>5,000 mg/kg (MRID 51868907 and 44408749), Acute Inhalation (LCso > 4.364
mg/L (could be waived based on results from MRID 44408709), and Dermal
Irritation (MRID 51868902). It is classified as toxicity category III for Eye
Irritation (i.e., caused moderate eye irritation that cleared in 7 days or less;
MRID 51868903) and is not a dermal sensitizer (MRID 51868906 and
44408752). MOEs were not calculated because there are no endpoints of concern
Page 57 of 118
-------
for the dermal route of exposure. The entire chronic toxicity database was
generated using dermal studies, including developmental/reproductive and
chronic studies, and there were no systemic toxic effects identified that would be
relevant to humans. In the acute dermal study, the LCso, NOEL and NOAEL
were all greater than 2,000 mg/kg, and there was no evidence of dermal irritation
or dermal sensitization. The Agency concluded that there is reasonable certainty
of no harm to populations or subpopulation (infants and children) from the use of
picaridin in insect repellent products applied to human skin.
(b) What is the nature of the risks to subjects of the proposed research?
Risks to test subjects include the risk of exposure to field mosquitoes and
mosquito-borne pathogens, the risk of exposure to the test materials, risks related
to receiving an unexpected result on a pregnancy test, health risks associated with
sitting and standing outdoors in areas with elevated temperature and humidity for
several hours, 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 spray product contains 10% PMD. Primary dermal irritation study (MRID
44438704) using 98.3 % pure PMD shows the test material to be non-irritating
after 4 hours of exposure. Erythema cleared in 72 hours for 4 out of 6 rabbits. No
traces of irritation were observed after 7 days. Acute dermal toxicity study using
98% PMD (MRID 44438702) applied at a dose of 5,000 mg/kg reports LCso >
5,000 mg/kg. The dose applied for testing is 1 g/600 cm2/day, lower than the
NOAEL = 1,000 mg/kg/day, and the LOAEL = 3,000 mg/kg/day of 98.3 % PMD.
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. These values
are based on a 90-Day dermal study in rats (MRID 44438710) that tested PMD
(98.3 % pure) at increasing doses of 0, 1,000 and 3,000 mg/kg/day.
The active ingredient picaridin is classified as toxicity category IV for Acute Oral
(LDso >5,000 mg/kg (MRID 51868905 and 44408748), Acute Dermal (LDso
>5,000 mg/kg (MRID 51868907 and 44408749), Acute Inhalation (LCso > 4.364
mg/L (could be waived based on results from MRID 44408709), and Dermal
Irritation (MRID 51868902). It is classified as toxicity category III for Eye
Irritation (i.e., caused moderate eye irritation that cleared in 7 days or less; MRID
51868903) and is not a dermal sensitizer (MRID 51868906 and 44408752).
MOEs were not calculated because there are no endpoints of concern for the
dermal route of exposure. The entire chronic toxicity database was generated
using dermal studies, including developmental/reproductive and chronic studies,
and there were no systemic toxic effects identified that would be relevant to
humans. In the acute dermal study, the LCso, NOEL and NOAEL were all greater
than 2,000 mg/kg, and there was no evidence of dermal irritation or dermal
Page 58 of 118
-------
sensitization. The Agency concluded that there is reasonable certainty of no harm
to populations or subpopulation (infants and children) from the use of picaridin in
insect repellent products applied to human skin.
(d) What is the probability of each risk associated with the research? How
was this probability estimated?
No numerical probability is estimated, but risks have a low probability of
occurrence. Practical steps to minimize test subject risks are described in the
protocol.
4.2 Risk minimization
(a) What specific steps are proposed to minimize risks to subjects?
The protocol outlines the risks and how they will be minimized on pp. 20-22.
These include:
• Employing the eligibility criteria to minimize the risk of adverse reaction
to the test substance and mosquito bites
• Monitoring the field sites in advance of the test days and coordinating
with state and local agencies to confirm that no mosquito-borne disease is
present at the test site
• Training subjects on the proper use of an aspirator to capture mosquitoes
that land before they can bite
• Informing subjects about the symptoms of mosquito-vectored diseases so
they can be identified, and subjects can seek medical attention if they
experience them
• Confirming the attractiveness of individuals to mosquitoes prior to their
participation in the field test to avoid exposing subjects to the risks of
mosquitoes in the field if they are not qualified to participate in the
research
• Protecting the privacy of female subjects during pregnancy testing
• Maintaining the confidentiality of subjects' identities
• Providing a shaded area, seating, cool drinks, and snacks for the subjects
to use between exposure periods
• Providing common over-the-counter first aid items (bandages, antiseptics,
antihistamines) to subjects upon request
• Providing qualified first aid assistance and transporting injured/ill
subjects to the closest facility for medical attention
The proposed risk mitigation measures seem reasonable, though incomplete. EPA
has recommended additional steps that should be taken to effectively minimize
the identified risks to subjects in this study.
(b) What stopping rules are proposed in the protocol?
Page 59 of 118
-------
• Each subjects' participation will stop when they experience FCL.
• "If adequate landing pressure on the control subjects is not achieved
during 3 consecutive exposure periods, the test will be stopped for all test
subjects" (p. 29).
• "If adverse reactions are observed during the test, the test subject will be
removedfrom the test immediately" (p. 30).
• "The Study Director or PI will stop, and if needed reschedule, the test
day if temperatures exceed 100°F, or other unanticipated weather arises
that poses unsafe conditions to remain outdoors" (p. 30).
(c) How does the protocol provide for medical management of potential
illness or injury to subjects?
Medical management will be provided through a combination of first aid supplies
available at the field site, as well as transportation to the closest hospital in the
event of a serious incident. Additionally, subjects will be instructed to monitor for
symptoms of mosquito-borne illness and to seek medical attention in the event
they experience such symptoms.
(d) 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?
The protocol does not specify a period for post-exposure monitoring or
follow-up. The EPA has proposed including a provision requiring the Study
Director to contact each subject after their participation at an interval that is
of long enough duration to discover any adverse events that might occur.
The protocol includes a provision for the Study Director to notify subjects if
diseases are identified in the mosquitoes collected in the field during each
day of the field testing.
(e) How and by whom will medical care for research-related injuries to
subjects be paid?
The consent form notes "In the unlikely event that you are injured as a result
of your participation in this study, medical care will be made immediately
available. The sponsor will reimburse you for the costs of this care." (p. 13)
The EPA has recommended that this information and more details about the
process for reimbursement be included in the protocol.
5. Benefits
Page 60 of 118
-------
(a) What benefits of the proposed research, if any, would accrue to individual
subjects?
No benefits would accrue to the subjects.
(b) What benefits to society are anticipated from the information likely
to be gained through the research?
Depending on the results of the research, it may benefit society by generating
reliable repellency efficacy data that could be used by the EPA to support
registration of insect repellent products containing the ingredients tested.
Registration of effective repellent products could lead to fewer mosquito bites and
reduced incidents of vector-borne illnesses.
(c) How would societal benefits be distributed? Who would benefit from
the proposed research?
Users of skin-applied repellents would benefit from the results of this research
through availability of effective products.
The Sponsor would benefit through registration of the products, which would
allow sale and distribution within the United States.
(d) What is the likelihood that each identified societal benefits would be
realized?
If the protocol is modified to address adequately all of the EPA's
recommendations, then it is likely the study will meet applicable scientific and
ethical standards, could be used to support registration of the product, and would
result in the potential sale of additional skin-applied repellent products to users.
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 protocol describes a procedure for generating scientifically sound data and
includes measures to limit the risks to subjects participating in the study. With the
risk mitigation measures put in place and provided that the EPA's comments
addressed, the remaining risk to subjects is low and reasonable in light of the
potential benefits of the data to society.
7. Independent Ethics Review
(a) What IRB reviewed the proposed research?
Page 61 of 118
-------
WCG Institutional Review Board IRB
(b) Is this IRB independent of the investigators and sponsors of the research?
Yes
(c) Is this IRB registered with OHRP?
Yes
(d) Is this IRB accredited? If so, by whom?
Yes, WCG IRB is accredited by the Association for the Accreditation of Human
Research Protection Programs.
(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
(e) What standard(s) of ethical conduct would govern the work?
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 the EPA under the pesticide laws. 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, with the EPA's recommendation to amend the protocol to ensure that each
subject will automatically receive a signed copy of the consent form incorporated.
(c) Do the informed consent materials meet the requirements of 40 CFR
26.1116, including adequate characterization of the risks and discomforts to
Page 62 of 118
-------
subjects from participation in the research, the potential benefits to the
subject or others, and the right to withdraw from the research?
With the EPA's recommendations regarding the consent form, as well as all
amendments to the protocol that would impact the consent process or
documentation addressed, the informed consent materials are likely to meet the
requirements of 40 CFR 26.1116.
(d) What is the literacy rate in English or other languages among the intended
research subjects?
All subjects are required to read and understand English.
(e) What measures are proposed to overcome language differences, if any,
between investigators and subjects?
None.
(f) What measures are proposed to ensure subject comprehension of risks and
discomforts?
The protocol does not include measures to ensure subject comprehension of the
risks and discomforts of participating or other aspects of their participation. The
EPA has made a recommendation for measures to be included in the protocol.
(g) What specific procedure will be followed to inform prospective subjects
and to seek and obtain their consent?
"Prior to participating in any aspect of the test, each potential subject who has
expressed interest in participation in the study and has met the
inclusion/exclusion criteria will be asked to review and be willing to sign an ICD.
Only the subjects that are willing to sign the ICD will be scheduledfor a training
date and any test date(s). Only the subjects that sign the ICD will participate in
the study. Subjects will be informed that they may withdraw from testing at any
time for any reason.
"7.4.1. At the training session the Study Director or Principal Investigator (PI)
will provide copies of the ICD, review the inclusion/exclusion criteria with the
participants and ask if they have any questions. Any questions will be answered.
"7.4.2. If a subject still wishes to enroll in the study and meets the criteria, he or
she will be asked to sign the ICD. His or her signature will be witnessed by the
Study Director or PI. A copy of the signed ICD will be offered to each subject
(p. 19)
Page 63 of 118
-------
(h) What measures are proposed to ensure fully voluntary participation
and to avoid coercion or undue influence?
Subjects will be informed multiple times throughout the process that they are free
to withdraw at any time without forfeiting any compensation or benefit to which
they are entitled. Employees of the sponsor and employees' immediate family
members are excluded from participation to avoid the perception of pressure to
participate in the study. The EPA recommends also excluding employees of the
recruitment firms and immediate family members of the employees for the same
reason.
9. Respect for Subjects
(a) How will information about prospective and enrolled subjects be
managed to ensure their privacy?
"• Each subject will be assigned a code number. Only subjects' code numbers will
appear on data sheets. The subjects' names will not appear in the report.
"• The report (as well as all study-related records) will not be publicly available
and will be kept as confidential as possible under local, state, and federal laws
and regulations. The study results generated following this protocol are not
intendedfor publication; however, if any of the study-related data are published,
subjects' identities will remain confidential.
"• The study records will be maintained at the study site in locked cabinets and
electronic files kept on a password-protected computer server.
"• No one outside SCJ, the study staff, the recruitment firm, the IRB, or certain
governmental agencies (such as USEPA) will have access to subjects 'personal
information." (pp. 21-22)
(b) How will subjects be informed of their freedom to withdraw from the
research at any time without penalty?
During the recruitment, consent process, training on aspirator use and
attractiveness testing, and on each study day, subjects will be reminded of their
freedom to withdraw at any time without penalty.
(c) How will subjects who decline to participate or who withdraw from the
research be dealt with?
Candidates who decline to participate in the research do not have provide consent.
Subjects who withdraw from the research will be compensated based on the
length of their participation prior to the withdrawal.
Page 64 of 118
-------
The EPA recommends that the protocol include specific provisions for how the
data from withdrawing subjects will be handled.
Page 65 of 118
-------
ATTACHMENT 2
Ethics Review Completeness Checklists
The following checklists are public documents. They are used by EPA in reviewing proposed
protocols for third-party research involving intentional exposure of human subjects to a
pesticide, with the intention of submitting the resulting data to EPA under FIFRA. These
checklists only address ethical requirements and do not address the scientific integrity of the
proposed study.
Checklist Associated with 40 CFR 26.1125
Submission of proposed human research for EPA review
Requirement
Y/N
Comments/Page Refs
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, and
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; and
a written summary of the discussion of controverted issues and their
resolution.
Y
Provided to EPA directly by
WCG IRB
(3) Records of continuing review activities, including the rationale for
conducting continuing review of research that otherwise would not require
continuing review as described in §26.1109(f)(1).
N/A
(4) Copies of all correspondence between the IRB and the investigators.
Y
(5) A list of IRB members in the same detail as described in
§26.1108(a)(2).
Y
(6) Written procedures for the IRB in the same detail as described in
§26.1108(a)(3) and (4).
Y
Provided to EPA directly by
WCG IRB
(7) Statements of significant new findings provided to subjects, as
required by §26.1116(c)(5).
N/A
The following additional information, to the extent not already included. A discussion of:
(a)(1) The potential risks to human subjects
Y
(a)(2) The measures proposed to minimize risks to the human subjects
Y
(a)(3) The nature and magnitude of all expected benefits of such
research, and to whom they would accrue
Y
(a)(4) Alternative means of obtaining information comparable to what
would be collected through the proposed research; and
Y
(a)(5) The balance of risks and benefits of the proposed research.
Y
(b) All information for subjects and written informed consent agreements
as originally provided to the IRB, and as approved by the IRB.
Y
Page 66 of 118
-------
(c) Information about how subjects will be recruited, including any
advertisements proposed to be used.
Y
All recruitment tools and scripts
must be provided to EPA before
the research is initiated.
(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
(e) All correspondence between the IRB and the investigators or
sponsors.
Y
(f) Official notification to the sponsor or investigator, in accordance with
the requirements of this subpart, that research involving human subjects
has been reviewed and approved by an IRB.
Y
Page 67 of 118
-------
Checklist Associated with 40 CFR §26.1116
General requirements for informed consent of human subjects
Criterion
Y/N
Comment/Page Reference
Consent Process - 40 CFR 26.1116(a)
(1) Before involving a human subject in research covered by this subpart,
an investigator shall obtain the legally effective informed consent of the
subject.
Y
(2) An investigator shall seek informed consent only under circumstances
that provide the prospective subject sufficient opportunity to discuss and
consider whether or not to participate and that minimize the possibility of
coercion or undue influence.
Y
Provided EPA's
recommendations are addressed.
(3) The information that is given to the subject shall be in language
understandable to the subject.
Y
(4) The prospective subject must be provided with the information that a
reasonable person would want to have in order to make an informed
decision about whether to participate, and an opportunity to discuss that
information.
Y
Provided EPA's
recommendations are addressed.
(5) (i) Informed consent must begin with a concise and focused
presentation of the key information that is most likely to assist a
prospective subject in understanding the reasons why one might or might
not want to participate in the research. This part of the informed consent
must be organized and presented in a way that facilitates comprehension,
(ii) Informed consent as a whole must present information in sufficient
detail relating to the research and must be organized and presented in a
way that does not merely provide lists of isolated facts, but rather
facilitates the prospective subject's understanding of the reasons why one
might or might not want to participate.
Y
Provided EPA's
recommendations are addressed.
(6) No informed consent may include any exculpatory language through
which the subject 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
Basic Elements of Informed Consent - 40 CFR 26.1116(b)
In seeking informed consent the following information shall be provided to each subject:
(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 that are experimental
Y
Provided EPA's
recommendations are addressed.
(2) A description of any reasonably foreseeable risks or discomforts to the
subject
Y
Provided EPA's
recommendations are addressed.
(3) A description of any benefits to the subject or to others that may
reasonably be expected from the research
Y
(4) A disclosure of appropriate alternative procedures or courses of
treatment, if any, that might be advantageous to the subject
N/A
(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
Provided EPA's
recommendations are addressed.
(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
Page 68 of 118
-------
(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 any time without
penalty or loss of benefits to which the subject is otherwise entitled; and
Y
(9) One of the following statements about any research that involves the
collection of identifiable private information or identifiable biospecimens:
(i) A statement that identifiers might be removed from the identifiable
private information or identifiable biospecimens and that, after such
removal, the information or biospecimens could be used for future
research studies or distributed to another investigator for future
research studies without additional informed consent from the subject, if
this might be a possibility; or
(ii) A statement that the subject's information or biospecimens collected
as part of the research, even if identifiers are removed, will not be used
or distributed for future research studies.
N
To be added per EPA's
recommendation.
Additional elements of informed consent - 40 CFR 26.1116(c)
One or more of the following elements of information, when appropriate, shall also be provided to each subject
(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) that are currently unforeseeable;
Y
(2) Anticipated circumstances under which the subject's participation may
be terminated by the investigator without regard to the subject's consent;
Y
(3) Any additional costs to the subject that may result from participation in
the research;
N/A
(4) The consequences of a subject's decision to withdraw from the
research and procedures for orderly termination of participation by the
subject;
Y
Provided EPA's
recommendations are addressed.
(5) A statement that significant new findings developed during the course
of the research that may relate to the subject's willingness to continue
participation will be provided to the subject;
N/A
(6) The approximate number of subjects involved in the study;
Y
(7) A statement that the subject's biospecimens (even if identifiers are
removed) may be used for commercial profit and whether the subject will
or will not share in this commercial profit;
N/A
(8) A statement regarding whether clinically relevant research results,
including individual research results, will be disclosed to subjects, and if
so, under what conditions; and
N/A
(9) For research involving biospecimens, whether the research will (if
known) or might include whole genome sequencing (i.e., sequencing of a
human germline or somatic specimen with the intent to generate the
genome or exome sequence of that specimen).
N/A
(h) 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
Provided EPA's
recommendations are addressed.
Page 69 of 118
-------
Checklist associated with 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 (including in an electronic
format) by the subject. A written copy shall be given to the subject.
Y
Provided EPA's
recommendations are
addressed.
(b) The informed consent form may be either of the following:
(1) A written informed consent form that meets the requirements of
§26.1116. The investigator shall give the subject adequate opportunity
to read the informed consent form before it is signed; alternatively, this
form may be read to the subject.
Y
Provided EPA's
recommendations are
addressed.
(2) A short form written informed consent form stating that the elements
of informed consent required by §26.1116 have been presented orally to
the subject, and that the key information required by §26.1116(a)(5)(i)
was presented first to the subject, before other information, if any, was
provided. The IRB shall approve a written summary of what is to be said
to the subject. When this method is used, there shall be a witness to the
oral presentation. Only the short form itself is to be signed by the
subject. 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 must be given to the
subject, in addition to a copy of the short form.
N/A
Page 70 of 118
-------
ATTACHMENT 3
S.C. Johnson Statistical Analysis to Support Sample Size
Provided as separate file
Page 71 of 118
-------
ATTACHMENT 4
Power/Sample Size Calculation for Mosquito Repellency Studies
where Complete Protection Time is the Endpoint
Date: 11/20/2017
Objective
To determine the sample size N such that mosquito 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
-------
«X,K,X) = ff$K",irC*/l)"
10 x < 0
e x > 0,
0 x < 0
F(x, k, }.) = {!
and are illustrated in the associated plots in Figures 1 and 2 for some illustrative k and X values.
Parameterizing the Weibull distribution in terms of k and X is, however, not necessarily intuitive
with respect to studying - and judging — the efficacy of skin-applied mosquito repellents as
measured by CPT for individuals using the repellent in the field. Instead, it is more natural and
desirable to be able to express the efficacy of the repellent in terms of both the expected
precision of the estimated median CPT (mCPT) and in terms of the estimated variability of
mCPT in (or across) the population. More specifically: the testing of a given repellent in the field
should be able to generate a reasonably precise estimate of the mCPT that is expected to be
generally close to what a sizable fraction of the population would be expected to experience (or,
more accurately, a mCPT that only a small fraction of the population would ideally experience to
be much shorter).
Following the above logic, we define the precision of the CPT estimate — designated as "K" — as
follows:
K = 95% LCLmCPT/estimated mCPT
where: mCPT= 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 = CPTsth ° oiie /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 forjudging 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 X (scale) values. To do this, EPA developed an equation such that
interconversion between the standard (k (shape) and X (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:
Page 73 of 118
-------
P(CPT, k, I) = 1 -e-(cpm)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:
_(rnŁPT\K
P(mCPT,K,X) = l — e\ a > =0.5 (median)
(PSMRxmCPT\K
P(PSMR x mCPT,K,X) = 1 — e v A J = 0.05 (5th percentile)
Algebraically solving the equations above (see Appendix A for full derivation), we develop
expressions for k and X:
[ln(0.95)
k = In
ln(0.5)
mCPTK
± xIn —,
X = e« V ln(0.5)
/ ln(P5Mi?)
Table 1 below compares these two parameterizations for the example PDF and CDF distributions
shown in Figures 1 and 2, respectively, for the k and X parameterizations shown there,
illustrating the conversion to this new parameterization:
Table 1. Re-parameterization of Weibull Distribution Parameters from Traditional (k, X) to Revised (P5MR,
mCPT) for Example Weibull Distributions Appearing in Figures 1 and 2.
Parameterization Scheme
Description/Comments
Traditional
Revised
Scale (k)3
Shape (k)
mCPTb
P5MRcd
1
0.5
0.480453
0.005476
- k values of less than 1 indicate a failure 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 failure rate
over time possibly suggesting failure 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 failure 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
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 2, as a consequence, that all \=1 distributions intersect
at the 63.2 percentile.
b mCPT = [ln(2)*exp(K * 1 n(\))](1/K)
c P5MR = exp( ln( 1 n(0.9 5)/ln (0.5))/k)
d Note that as k increases, the P5MR value becomes larger, indicating that the values at the 5th percentile approaches 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.
An example of the (varied) kinds of distributional "shapes" associated with various
parameterizations is shown in Figure 2 as histograms of the CPT. More specifically, Figure 2
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
Page 74 of 118
-------
the P5MR ratio (P5MR= 0.2, 0.3, 0.4, and 0.6) for the (assumed) Weibull Distribution18. As seen
in Figure 2, 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 for the field exercise to achieve a
desired set of aims regarding precision around the estimate of the mCPT. These judgments are
based in part on available data and past experiences19 and in part on general thoughts regarding
consumer 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 greater than a given acceptable K (a
scalar which measures the precision of the estimates in estimating the mCPT). Such mosquito
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% LCLmCPT/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% LCLmCPT/estimated
mCPT > K (and the trial is considered to be a "success" in the power
calculation)
18Other simulations were performed for the lognormal, normal, and uniform distributions, with the latter one
(particularly) done as a form of sensitivity analysis but these are not discussed in this report; the simulation outputs,
however, are provided in Appendix D. Note that the power estimates for a given sample size from the Weibull and
Lognormal distributions are similar.
19See Appendix B for Weibull parameters fit to previous mosquito efficacy field data that the EPA has evaluated for a
similar design and experimental set-up. In general, the values found in these (prior) studies support the values
selected here to be used for the simulation
Page 75 of 118
-------
• 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, certerisparibus).
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 mosquito 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 and 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 Field Mosquitoes Repellent Study
In mosquito field repellency studies, test subjects are exposed in the field for 5-minute intervals
immediately following product application and then for 5 minutes every 30 minutes until a "first
confirmed landing" occurs. For subjects who receive confirmed landings, the CPTs are set as 0 if
the first confirmed landing occurs during the first 5 minutes after application of the repellent;
otherwise, the CPTs are rounded down to the nearest half hour (i.e., the starting time of the
exposure period in which the first confirmed landing occurs). For those subjects for which there
are no confirmed landings 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 (Computer) Simulation Procedure:
To simulate the field 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 < 5, 6-35, 36-65, 66-95, ... 576-605 minutes, the CPTs
are set to be 0-, 0.5-, 1-, 1.5 hours... 10 hours, respectively, to simulate the study design in which
each study participant is exposed for 5 of every 30 minutes until the first confirmed mosquito
landing. If the randomly generated CPTs are greater than 10 hours (or 605 minutes), they are
considered in the calculation to be (right) censored at 10 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% LCLmCPT/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.
Page 76 of 118
-------
The process described in previous paragraph is then repeated for each combination of different
mCPT = 2, 4, 6, and 8 hours; P5MR = 0.2, 0.4, 0.5, 0.6, 0.7, and 0.8; sample size per dataset =
10, 11,12... 20; and the lowest acceptable K = 0.6, 0.7, and 0.8; all assuming that CPT follows
a Weibull distribution20.
20 Such calculations were similarly done for the lognormal distribution, normal distribution, and uniform
distribution, but are not discussed further in this report. The SAS output from these calculations and various
associated tables and graphs, however, is shown in Appendix D for completeness.
Page 77 of 118
-------
Results of Simulation
Tables 2, 3, and 4 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, respectively. These are shown for
various values of mCPT (ranging from 2 to 8 hours), P5MR (ranging from 0.2 to 0.8), and
Sample Size (ranging from 10 to 20). 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 a 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.
Figures 4, 5, and 6 present visually the same results in Tables 2, 3, and 4 (as power curves rather
than tables).
As can be seen within each Table or Figure, 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). 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 Figures or 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.
The SAS Code used to generate the simulated data and the associated tables and graphs are
presented in Appendix C. Note - as described earlier - those simulations were also performed for
the lognormal, normal, and uniform distributions, in part to serve as a sensitivity analysis and
these are presented in the Appendix D for completeness but are not discussed further here.
Page 78 of 118
-------
Figure 1. Probability Density Function (PDF) for Weibull Plot with X (scale) =1 and k (shape)
ranging from 0.5 to 5.
Weibull(2-Parameter) Probability density function
0.0 0.5 1.0 1.5 2.0 2.5
Figure 2. Cumulative Distribution Function (CDF) for Above Weibull PDF.
Weibull(2-Parameter) Cumulative distribution function
0.0 0.5 1.0 1.5 2.0 2.5
Page 79 of 118
-------
Figure 3: Histograms of CPT distributions for various CPTs and P5MRs (assume CPTs are
Weibull distributions).
weibull distribution: median=2
P5MR= 0.2
TTTTTTTrmrm
I ^ I ^ I I I I I i ^ ^ i i * * * '
TllTTlTTTl-r^_
CPT
weibull distribution: median=4
6-
4-
2-
6
4-
2
6
4-
2
6-
4-
2-
6-
4-
2-
P5MR= 0.2
tm LQHmJ
^i-mifTTT.
111
TllTTTTri-i-rT^—
Thri-i^-
_=ti=dj[
I i I —i 1 1 r
23456789 10
CPT
weibull distribution: median=6
weibull distribution: median=8
P5MR = 0.2
TrrrrrrriTTTTTTTTTTTTT^
irmrnrTm^
P5MR= 0.4
iTTTTTTrh^,
P5MR= 0.6
"I i i i i i i r
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
CPT
6-
4-
2-
P5MR = 0.2
P5MR = 0.4
~iTTTTTTrh^_
P5MR = 0.6
w
~i i I i I i i i i I r
0 1 2 3 4 5 6 7 8 9 10111213141516171819 20
CPT
Page 80 of 118
-------
Table 2: Results of power analysis when the lowest acceptable ratio 95% LCLmCpT/mCPT = 0.6
(Weibull distribution)
Median
(hours)
P5MR
Sample size
10
11
12
13
14
15
16
17
18
19
20
2
0.2
0.071
0.291
0.207
0.473
0.362
0.369
0.502
0.494
0.637
0.626
0.521
0.4
0.297
0.691
0.594
0.841
0.804
0.779
0.893
0.898
0.939
0.945
0.932
0.5
0.498
0.850
0.802
0.942
0.938
0.921
0.968
0.977
0.964
0.982
0.986
0.6
0.733
0.949
0.943
0.962
0.971
0.955
0.954
0.979
0.915
0.951
0.971
0.7
0.893
0.945
0.955
0.875
0.918
0.852
0.855
0.893
0.810
0.859
0.886
0.8
0.819
0.786
0.826
0.666
0.734
0.591
0.637
0.708
0.558
0.632
0.689
4
0.2
0.043
0.208
0.146
0.356
0.289
0.254
0.432
0.380
0.567
0.516
0.435
0.4
0.241
0.595
0.521
0.783
0.737
0.709
0.849
0.842
0.930
0.920
0.884
0.5
0.412
0.795
0.730
0.921
0.901
0.888
0.956
0.964
0.986
0.988
0.973
0.6
0.648
0.938
0.899
0.987
0.980
0.976
0.995
0.997
0.997
0.999
0.998
0.7
0.869
0.988
0.986
0.993
0.995
0.994
0.992
0.998
0.977
0.992
0.996
0.8
0.975
0.982
0.987
0.948
0.970
0.949
0.934
0.968
0.887
0.932
0.954
6
0.2
0.075
0.204
0.153
0.339
0.280
0.252
0.426
0.369
0.557
0.490
0.424
0.4
0.227
0.572
0.504
0.759
0.743
0.689
0.851
0.826
0.929
0.916
0.885
0.5
0.408
0.779
0.729
0.914
0.905
0.873
0.963
0.958
0.987
0.981
0.978
0.6
0.645
0.925
0.906
0.984
0.980
0.977
0.997
0.997
1.000
0.999
0.999
0.7
0.874
0.990
0.988
0.998
0.999
0.999
1.000
1.000
0.999
1.000
1.000
0.8
0.986
0.998
0.999
0.993
0.995
0.994
0.990
0.997
0.975
0.989
0.995
8
0.2
0.323
0.346
0.362
0.457
0.443
0.361
0.537
0.453
0.636
0.564
0.522
0.4
0.314
0.586
0.552
0.769
0.753
0.700
0.858
0.836
0.934
0.919
0.891
0.5
0.421
0.779
0.732
0.914
0.904
0.875
0.960
0.956
0.989
0.985
0.979
0.6
0.638
0.927
0.906
0.983
0.979
0.974
0.997
0.997
1.000
0.999
1.000
0.7
0.874
0.990
0.989
0.999
1.000
0.999
1.000
1.000
1.000
1.000
1.000
0.8
0.985
0.999
1.000
0.997
1.000
0.999
0.998
1.000
0.994
0.998
0.999
NOTE: Yellow indicates power > 0.8; orange indicates power > 0.9; blue indicates unusual power
when median complete protection time = 2 hours and P5MR = 0.8.
Page 81 of 118
-------
Table 3: Results of power analysis when the lowest acceptable ratio 95% LCLmCpT/mCPT = 0.7
(Weibull distribution)
Median
(hours)
P5MR
Sample size
10
11
12
13
14
15
16
17
18
19
20
2
0.2
0.022
0.096
0.066
0.199
0.121
0.128
0.217
0.211
0.304
0.299
0.222
0.4
0.132
0.415
0.299
0.607
0.484
0.522
0.634
0.677
0.751
0.767
0.668
0.5
0.267
0.633
0.516
0.789
0.697
0.748
0.803
0.851
0.868
0.895
0.838
0.6
0.476
0.813
0.732
0.881
0.845
0.864
0.885
0.919
0.893
0.926
0.918
0.7
0.694
0.895
0.876
0.861
0.889
0.847
0.850
0.888
0.799
0.837
0.888
0.8
0.768
0.780
0.821
0.673
0.750
0.591
0.652
0.699
0.566
0.622
0.694
4
0.2
0.016
0.075
0.053
0.166
0.109
0.088
0.190
0.171
0.276
0.245
0.177
0.4
0.103
0.332
0.267
0.517
0.452
0.402
0.620
0.555
0.752
0.681
0.638
0.5
0.210
0.525
0.468
0.715
0.685
0.624
0.830
0.776
0.914
0.866
0.848
0.6
0.402
0.736
0.714
0.886
0.880
0.833
0.955
0.923
0.979
0.966
0.969
0.7
0.673
0.914
0.917
0.971
0.975
0.962
0.987
0.986
0.982
0.988
0.995
0.8
0.927
0.970
0.987
0.945
0.971
0.946
0.931
0.955
0.892
0.922
0.958
6
0.2
0.047
0.083
0.066
0.158
0.105
0.083
0.174
0.150
0.247
0.218
0.149
0.4
0.079
0.294
0.225
0.473
0.387
0.356
0.556
0.507
0.690
0.636
0.566
0.5
0.172
0.483
0.406
0.679
0.622
0.573
0.779
0.735
0.887
0.841
0.806
0.6
0.335
0.697
0.649
0.861
0.851
0.804
0.938
0.909
0.977
0.963
0.956
0.7
0.607
0.894
0.885
0.975
0.970
0.958
0.994
0.989
0.997
0.996
0.997
0.8
0.897
0.987
0.992
0.988
0.995
0.993
0.989
0.993
0.978
0.987
0.995
8
0.2
0.309
0.234
0.297
0.289
0.320
0.210
0.347
0.251
0.392
0.306
0.306
0.4
0.180
0.321
0.294
0.497
0.435
0.379
0.598
0.521
0.726
0.654
0.592
0.5
0.206
0.499
0.439
0.692
0.645
0.603
0.804
0.762
0.904
0.867
0.830
0.6
0.357
0.731
0.684
0.892
0.872
0.831
0.957
0.933
0.983
0.978
0.966
0.7
0.634
0.922
0.907
0.985
0.981
0.976
0.998
0.994
0.998
0.998
0.999
0.8
0.913
0.996
0.995
0.999
0.999
0.998
0.997
1.000
0.994
0.997
0.999
NOTE: Yellow indicates power > 0.8; orange indicates power > 0.9; blue indicates unusual power
when median complete protection time = 2 hours and P5MR = 0.8.
Page 82 of 118
-------
Table 4: Results of power analysis when the lowest acceptable ratio 95% LCLmCpT/mCPT = 0.8
(Weibull distribution)
Median
(hours)
P5MR
Sample size
10
11
12
13
14
15
16
17
18
19
20
2
0.2
0.007
0.032
0.027
0.064
0.044
0.027
0.079
0.061
0.115
0.088
0.071
0.4
0.035
0.119
0.103
0.219
0.188
0.140
0.275
0.227
0.355
0.287
0.258
0.5
0.068
0.210
0.185
0.339
0.297
0.251
0.418
0.364
0.506
0.460
0.425
0.6
0.133
0.341
0.306
0.496
0.473
0.407
0.594
0.560
0.678
0.670
0.641
0.7
0.251
0.561
0.532
0.679
0.692
0.598
0.753
0.753
0.749
0.775
0.809
0.8
0.455
0.696
0.728
0.654
0.728
0.562
0.648
0.694
0.565
0.620
0.692
4
0.2
0.004
0.026
0.012
0.053
0.027
0.022
0.054
0.045
0.085
0.065
0.042
0.4
0.026
0.093
0.080
0.186
0.151
0.103
0.254
0.182
0.346
0.245
0.228
0.5
0.060
0.170
0.165
0.315
0.297
0.202
0.436
0.315
0.546
0.413
0.414
0.6
0.135
0.317
0.320
0.494
0.499
0.374
0.650
0.529
0.760
0.643
0.639
0.7
0.295
0.548
0.565
0.726
0.754
0.651
0.863
0.784
0.914
0.854
0.873
0.8
0.619
0.828
0.867
0.884
0.923
0.864
0.913
0.922
0.886
0.909
0.947
6
0.2
0.038
0.033
0.027
0.055
0.037
0.027
0.053
0.033
0.076
0.058
0.039
0.4
0.022
0.098
0.072
0.206
0.135
0.115
0.234
0.196
0.341
0.289
0.214
0.5
0.054
0.205
0.154
0.364
0.281
0.248
0.438
0.382
0.567
0.493
0.425
0.6
0.133
0.383
0.335
0.572
0.525
0.473
0.694
0.626
0.812
0.748
0.716
0.7
0.316
0.646
0.614
0.819
0.818
0.750
0.918
0.874
0.965
0.943
0.938
0.8
0.670
0.916
0.917
0.967
0.974
0.962
0.986
0.984
0.977
0.985
0.993
8
0.2
0.301
0.193
0.270
0.198
0.264
0.155
0.250
0.157
0.244
0.171
0.206
0.4
0.122
0.136
0.141
0.227
0.182
0.142
0.267
0.208
0.340
0.292
0.229
0.5
0.082
0.209
0.165
0.368
0.282
0.256
0.434
0.392
0.561
0.505
0.424
0.6
0.124
0.390
0.321
0.588
0.514
0.490
0.688
0.655
0.823
0.779
0.710
0.7
0.299
0.683
0.610
0.857
0.808
0.794
0.915
0.909
0.966
0.963
0.939
0.8
0.644
0.940
0.909
0.989
0.978
0.981
0.994
0.995
0.993
0.995
0.998
NOTE: Yellow indicates power > 0.8; orange indicates power > 0.9.
Page 83 of 118
-------
Figure 4: Power curves of study design when the lowest acceptable ratio 95% LCLmCPT/rnCPT = 0.6 (Weibull distributions).
Weibull median = 2 hours, K = 0.6
Weibull median = 4 hours, K = 0.6
|5%-tile/median ratio o 0.2 + Q.4 x 0.5 a 0.6 ~ 0.7 i 0.8 |
15%-tile/median ratio o 0.2 + Q.4 x Q.5 A 0.6 ~ 0.,' 0.S |
Weibull median = 6 hours, K = 0.6
Weibull median = 8 hours, K = 0.6
Page 84 of 118
-------
Figure 5: Power curves of study design when the lowest acceptable ratio 95% LCLmCPT/rnCPT = 0.7 (Weibull distributions).
Weibull median = 2 hours. K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a Q.6 ~ 0.7 t 0.8 |
Weibull median = 4 hours, K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 i 0.8 |
Weibull median = 6 hours, K = 0.7
Weibull median = 8 hours, K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 A Q.6 ~ 0.: 0.8 |
15%-tile/median ratio o 0.2 +0.4 x 0.5 A 0.6 ~ 0.,' 0.8 |
Page 85 of 118
-------
Figure 6: Power curves of study design when the lowest acceptable ratio 95% LCLmCPT/rnCPT = 0.8 (Weibull distributions).
Weibull median = 2 hours. K = 0.8
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a Q.6 ~ 0.7 t 0.8 |
Weibull median = 4 hours, K = 0.8
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 i 0.8 |
Weibull median = 6 hours, K = 0.8
Weibull median = 8 hours, K = 0.8
15%-tile/median ratio o 0.2 + 0.4 x 0.5 A Q.6 ~ 0.: 0.8 |
15%-tile/median ratio o 0.2 +0.4 x 0.5 A 0.6 ~ 0.,' 0.8 |
Page 86 of 118
-------
APPENDIX A
Re-parameterization of Standard Weibull Equation
Given the definition of PDF and CDF from first principles:
fmCPT\
= 1 - e \ A J =
Then:
P(mCPT ,k,X) = 1 — e
_(¦
P(PSMR x mCPT, k,A) = 1-e
!mCPT\K
0.5 (median)
P5MRxmCPT\K
fmCPT\
J = o.
,-(¦
P5MRxmCPT\K
5 (median)
e
and
/mCPT\K
= 0.95 (5th percentile)
= ln(0.5)
V A
/P5MRxmCPT\K
V A
Divide (2) by (1), we have:
P5MR x mCPT~,K
= ln(0.95)
mCPT
k = In
ln(0.95)
ln(0.5)
ln(0.95)
ln(0.5)
/ln(P5Mi?)
From (1):
mCPT\K
= — ln(0.5)
(mCPT\
KXln\—-—J = ln[—ln(0.5)]
In
/mCPT\ 1
T") =
In (mCPT) — ln(2) = — ln[—ln(0.5)]
K
1
ln(2) = In (mCPT) — — ln[—ln(0.5)]
K
1
= — [k In (mCPT) — ln[—ln(0.5)]]
K
= 0.05 (5th percentile)
(1)
(2)
(3)
-------
= — [ln(mCPTK)
K
mCPTh
ln[— ln(0.5)]]
A =
1
= - In
K
¦L r mCPTK
[ ln(0.s)
ln(0.5)J
(4)
So.
k = In
ln(0.95)
ln(0.5)
/ ln(P5Mi?)
± x In -
mCPTK
A =
-------
APPENDIX B
Product
Location
Sample
size
Est. mCPT
(95% CI)
Ratio of
95% LCL/est. mCPT
Est. Weibull
(shape k; scale /.)
Est.
P5MR
A
1
10
7.5 (4.0, 8.0)
0.53
6.602; 7.777
0.674
2
10
8.5 (4.5, 10.0)
0.53
5.855; 8.624
0.641
B
1
10
12.0 (6.0, 12.0)
0.50
4.311; 10.669
0.547
2
10
12.0 (8.5, 12.0)
0.71
10.424; 11.516
0.779
C
1
10
7.5 (4.0, 9.0)
0.53
4.430; 8.146
0.556
2
8
5.0 (2.5, 5.5)
0.50
5.318; 4.915
0.613
D
1
10
2.0 (1.5, 2.0)
0.75
7.004; 2.135
0.690
2
10
2.5 (1.0, 3.5)
0.40
3.557; 2.8970
0.481
E
1
10
8.25 (6.0, 10.0)
0.73
7.609; 8.733
0.710
2
10
8.0 (3.5, 8.5)
0.44
4.009; 7.442
0.522
-------
APPENDIX
SAS codes
c
* Programmer: James Nguyen, USEPA
* Project: Mosguito Repellency Studies
* Purpose: Power Analysis/sample size calculation
* Description:
* - distributions: Weibull, Normal, Lognormal, Uniform
* - create histograms of the distributions
* - SAS Procedures: PROC LIFETEST and PROC ICLIFETEST
* Review Date: 4/10/2017
options formdlim=n=" ps=90 ls=90 nonumber nodate;
libname MOS "C:\Users\JNguyen\Desktop\MOS";
%Macro dlstParam;
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, b];
a = MED*(0.5* P5MR - 0.05)/0.45; b = MED*2 - a;
end;
if upcase(Distribution) = "NORMAL" then do;
*normal = N(a,b);
a = MED; b = MED*(1-P5MR)/I.645;
end;
if upcase(Distribution) = "LOGNORMAL" then do;
* lognormal = exp(N(a,b));
a = log(MED); b = (log(MED)-log(MED*P5MR))/I.645;
end;
%Mend;title;
%Macro generate;
if upcase(Distribution)
if upcase(Distribution)
if upcase(Distribution)
if upcase(Distribution)
%Mend;
"WEIBULL" then CPT = 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 Histogram(MED=, P5MRS=, dist=, seed=);
ilet N=l;
fclet P5MR&N = %nrbguote(%scan(&P5MRS,&N, %str( )));
%do %while (&&P5MR&N /N=);
%let N=%eval(&N+1);
%let P5MR&N = %nrbguote(%scan(&P5MRS,&N, %str(
&end;
fclet N=%eval(&N-1);
Data Parameters;
MED = &MED;
%do i = 1 %to &N;
P5MR = & & P5MR&i;
P5 = MED* P5MR;
output;
-------
%end;
label MED = "P50";
run;
Data Parameters;
set Parameters;
Distribution = "&dist";
% di s fcParam;
data simmer;
call streaminit(&seed) ;
set parameters;
do i = 1 to 50000;
%grenerate;
output;
end; *i;
drop i a b;
title "&dist distribution: median=&MED";
Proc SGPANEL data = Simmer;
panelby P5MR/rows=&N;
Histogram CPT/binwidth=%sysevalf(2.5*&MED/50);
refline P5 /axis=x lineattrs=(pattern=l thickness=l color=blue);
refline MED/axis=x lineattrs=(pattern=l thickness=l color=blue);
colaxis values = (0 to %sysevalf(2.5*&MED) by 1);
run;
Proc datasets nolist; save sasmacr; run;quit;
%Mend;title;
iHistogram (MED=2, P5MRS=
% Histogram(ME D=4, P5MRS=
lHistogram{MED=6, P5MRS=
% Histogram(ME D=8, P5MRS=
I Histogram{MED=10, P5MRS=
0.2 0.3 0.4 0.5 0.6,
0.2 0.3 0.4 0.5 0.6,
0.2 0.3 0.4 0.5 0.6,
0.2 0.3 0.4 0.5 0.6,
0.2 0.3 0.4 0.5 0.6,
dist=weibull,
dist=weibull,
dist=weibull,
dist=weibull,
dist=weibull,
seed=279420)
seed=279420)
seed=279420)
seed=279420)
seed=279420)
%Macro Histograml(MED=, P5MR=, seed=);
Data Parameters;
MED = &MED;
P5MR = &P5MR;
P5 = MED* P5MR;
do i = 1 to 4;
if i = 1 then Distribution = "Lognormal''
if i = 2 then Distribution = "Normal";
if i = 3 then Distribution = "Uniform";
if i = 4 then Distribution = "Weibull";
% di s tParam;
output;
end;
label MED = "P50" P5MR="5%-tile/median"; drop i
data simmer;
call streaminit(&seed);
set parameters;
do i = 1 to 50000;
% generate;
output;
end; *i;
drop i a b;
title "median=&MED P5MR=&P5MR" ;
Proc SGPANEL data = Simmer;
panelby Distribution/rows=4;
-------
Histogram CPT/binwidth=%sysevalf(2.5*&MED/50);
refline P5 /axis=x lineattrs=(pattern=l thickness=l
refline MED/axis=x lineattrs=(pattern=l thickness=l
colaxis values = (0 to %sysevalf(2.5*&MED) by 1);
run;
Proc datasets nolist; save sasmacr; run;quit;
%Mend;title;
color=blue)
color=blue)
&Histogram!(ME D=2, P5MR=0.2, seed=279420)
I Histogram! (MED=2,
%Histogram!(ME D=2,
%Histogram!(ME D=2,
% Histogram!(ME D=2,
P5MR=0.4,
P5MR=0.5,
P5MR=0.6,
P5MR=0.7 ,
seed=279420)
seed=279420)
seed=279420)
seed=279420)
RT = 0; CPT= 0;
%Macro CPT;
CPT=CPT* 60;
if CPT <= 5 then do;
LT = 0;
end;
else if CPT >= &maxT*60 then do;
LT = &maxT* 60; RT=.; CPT=&maxT*60; censor = 1;
end;
else do;
LT = 30*floor((CPT-5)/30)+5; RT = 30*ceil((CPT-5)/30);
censor = 0;
end;
censor = 0;
CPT = RT;
CPT = CPT/60;
LT = LT/60;
RT = RT/60;
%Mend;title;
%Macro power;
ods select none;
%if &censor=right %then %do;
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;
%end;
%if &censor=interval %then %do;
ods output quartiles=MPT;
Proc iclifetest data = simmer(keep=MED P5MR N Sim LT RT) method=turnbull
impute(seed=1234);
by MED P5MR N Sim;
time (LT, RT);
run;
%end;
ods select default;
Proc datasets nolist; delete simmer; run;quit;
Data MPT;
set MPT;
if percent = 50;
power = (LowerLimit >= &K*Estimate) ;
%if &censor=right %then %do; Censor = "right";%end;
%if &censor=interval %then %do; Censor = "interval"; %end;
Proc SQL;
create table &dist&MED as
select Censor, MED, P5MR, N, avg(Power) as Power
from MPT
group by Censor, MED, P5MR, N;
quit;
-------
%Mend;title;
%Macro Mosquito(med=, P5MRS=, nmin=,nmax=,maxT=,K=,dist=,censor=,NSim=, seed=);
%let N=1;
%let P5MR&N = %nrbquote(%scan(&P5MRS,&N, %str( )));
%do %while (&&P5MR&N /N=);
%let N=%eval(&N+1);
%let P5MR&N = %nrbquote(%scan(&P5MRS,&N, %str( )));
%end;
%let N=%eval(&N-1);
%do 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";
% di s tParam;
run;
data simmer;
call streaminit(&seed);
set Parameters;
do N = &Nmin to &Nmax;
do Sim = 1 to &NSim;
do ID = 1 to N;
%generate;
output;
end; *ID;
end; *Sim;
end; *N;
drop a b;
run;
Data Simmer;
set Simmer;
%CPT;
run;
%power;
Data All_&dist&MED;
set All_&dist&MED &dist&MED;
run;
Proc datasets nolist; delete Parameters simmer MPT &dist&MED; quit;
%end;
Data MOS.&distcensor._MED&MED._K%sysevalf(100* & K);
set All_&dist&MED;
run;
Proc datasets nolist; save sasmacr; run;quit;
%Mend;
dm loq 'clear';%Mbsguifco(med=2, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8, nmin=10, nmax=20, maxT=10,
K = 0.6, dist= weibull, censor=riqht, NSim=4000, seed=561);
dm loq 'clear';%Mbsguifco(med=4, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8, nmin=10, nmax=20, maxT=10,
K = 0.6, dist= weibull, censor=riqht, NSim=4000, seed=561);
-------
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.7
dm log
K = 0.7
dm log
K = 0.7
dm log
K = 0.7
dm log
K = 0.8
dm log
K = 0.8
dm log
K = 0.8
dm log
K = 0.8
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.7
dm log
K = 0.7
dm log
K = 0.7
dm log
K = 0.7
dm log
K = 0.8
dm log
K = 0.8
dm log
K = 0.8
dm log
K = 0.8
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.7
dm log
K = 0.7
dm log
K = 0.7
dm log
K = 0.7
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
clear
dist
%Mosqu±to(med=6, P5MRS=
weibull, censor=right,
%Mosquito{med=8, P5MRS=
weibull, censor=right,
0.2 0.4 0.5 0.6 0.7 0.8
NSim=4000, seed=561);
0.2 0.4 0.5 0.6 0.7 0.8
NSim=4000, seed=561);
%Mosc[iilto (med=2, P5MRS=0 . 2 0.4 0.5 0.6 0.7 0.8
weibull, censor=right, NSim=4000, seed=352);
%Mosquito{med=4, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8
weibull, censor=right, NSim=4000, seed=352);
%Mosquito{med=6, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8
weibull, censor=right, NSim=4000, seed=352);
%Mosquito{med=8, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8
weibull, censor=right, NSim=4000, seed=352);
%Mosquito{med=2, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8
weibull, censor=right, NSim=4000, seed=352);
%Mosquito{med=4, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8
weibull, censor=right, NSim=4000, seed=352);
%Mosquito{med=6, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8
weibull, censor=right, NSim=4000, seed=352);
%Mosquito{med=8, P5MRS=0.2 0.4 0.5 0.6 0.7 0.8
weibull, censor=right, NSim=4000, seed=352);
%Mosquito{med=2, P5MRS=0.
Lognormal, censor=right,
%Moscpi±to (med=4 , P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=6, P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=8, P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=2, P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=4 , P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=6, P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=8, P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=2, P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=4 , P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=6, P5MRS=0 .
Lognormal, censor=right,
%Moscpi±to (med=8, P5MRS=0 .
Lognormal, censor=right,
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=561)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=561)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=561)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=561)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=352)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=352)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=352)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=352)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=352)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=352)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=352)
2 0.4 0.5 0.6 0.7 0.
NSim=4000, seed=352)
%Moscpiito (med=2, P5MRS=0 . 2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=561)
%Mosgu±to(med=4, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=561)
%Mosgu±to(med=6, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=561)
%Mosgu±to(med=8, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=561)
%Mosquito{med=2, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=352)
%Mosgu±to(med=4, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=352)
%Mosgu±to(med=6, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=352)
%Mosgu±to(med=8, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=352)
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmin=10
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
nmax=20
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
maxT=10,
-------
dm log 'clear'
K = 0.8, dist=
dm log 'clear'
K = 0.8, dist=
dm log 'clear'
K = 0.8, dist=
dm log 'clear'
K = 0.8, dist=
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.6
dm log
K = 0.6
' clear'
. dist=
' clear'
. dist=
' clear'
. dist=
' clear'
. dist=
dm log 'clear'
K = 0.7, dist=
dm log 'clear'
K = 0.7, dist=
dm log 'clear'
K = 0.7, dist=
dm log 'clear'
K = 0.7, dist=
dm log 'clear'
K = 0.8, dist=
dm log 'clear'
K = 0.8, dist=
dm log 'clear'
K = 0.8, dist=
dm log 'clear'
K = 0.8, dist=
;%Mosqu±to(med=2, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=352);
;%M.oscpilto(med=4, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=352);
;%M.oscpilto(med=6, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=352);
; %M.oscpilto(med=8, P5MRS=0.2 0.4 0.5 0.6 0.7
Normal, censor=right, NSim=4000, seed=352);
; %Mosc[iilto (med=2, P5MRS=
Uniform, censor=right,
' %Mosc[U.ifco(med=4, P5MRS=
Uniform, censor=right,
' %MosqTiifco (med=6, P5MRS=
0.2 0.4 0.5 0.6 0.7
NSim=4000, seed=561)
0.2 0.4 0.5 0.6 0.7
NSim=4000, seed=561)
0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=561!
;%M.oscpilto(med=8, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=561]
;%M.oscpilto(med=2, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=352]
;%M.oscpilto(med=4, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=352]
; %M.oscpilto(med=6, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=352]
;%M.oscpilto(med=8, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=352]
;%M.oscpilto(med=2, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=352]
;%M.oscpilto(med=4, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=352]
;%M.oscpilto(med=6, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=352]
; %M.oscpilto(med=8, P5MRS=0.2 0.4 0.5 0.6 0.7
Uniform, censor=right, NSim=4000, seed=352]
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10, nmax=20, maxT=10
nmin=10,
nmin=10,
nmin=10,
nmin=10,
nmax=20,
nmax=20,
nmax=20,
nmax=20,
maxT=10
maxT=10
maxT=10
maxT=10
/*
dm log 'clear'
K = 0.7, dist=
dm log 'clear'
K = 0.7, dist=
dm log 'clear'
K = 0.7, dist=
dm log 'clear'
K = 0.7, dist=
*/
:%Mosguito(med=2, P5MRS=0.2 0.4 0.5 0.6 0.7 O.E
Weibull, censor=interval, NSim=4000, seed=352)
:%Mosguito(med=4, P5MRS=0.2 0.4 0.5 0.6 0.7 O.E
Weibull, censor=interval, NSim=4000, seed=352)
:%Mosguito(med=6, P5MRS=0.2 0.4 0.5 0.6 0.7 O.E
Weibull, censor=interval, NSim=4000, seed=352)
:%Mosguito(med=8, P5MRS=0.2 0.4 0.5 0.6 0.7 O.E
Weibull, censor=interval, NSim=4000, seed=352)
nmin=l0,
nmin=l0,
nmin=l0,
nmin=l0,
nmax=2 0,
nmax=2 0,
nmax=2 0,
nmax=2 0,
maxT=10
maxT=10
maxT=10
maxT=10
:> Create Figures and Print Results;
libname MOS "C:\Users\JNguyen\Desktop\MOS";
%let folder=C:\Users\JNguyen\Desktop\MOS;
%Macro SGPLOT(distribution=, K=);
title "&distribution median = 2 hours, K = O.&K";
Proc SGPLOT data = MOS.&distribution._right_med2_k&k.0;
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 "&distribution median = 4 hours, K = O.&K";
Proc SGPLOT data = MOS.&distribution._right_med4_k&k.0;
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 "&distribution median = 6 hours, K = O.&K";
Proc SGPLOT data = MOS.&distribution._right_med6_k&k.0;
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 "&distribution median = 8 hours, K = O.&K";
Proc SGPLOT data = MOS.&distribution._right_med8_k&k.0;
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;
%Mend;
%Macro print(distribution^ K=);
data &distribution._K&K;
set MOS.&distribution._right_med2_k&k.0
MOS.&distribution._right_med4_k&k.0
MOS.&distribution._right_med6_k&k.0
MOS.&distribution._right_med8_k&k.0;
run;
Proc transpose data = &distribution._K&K out = &distribution._K&K(drop=_NAME_);
by MED P5MR;
ID N;
var Power;
run;
title "Sdistribution K=0.&K.0" ;
Proc print data = &distribution._K&K noobs label; format 6.3; run;
%mend;
%SGPŁOT(distribution=Weibull, K=6);
%SGPLOT{distribution=Weibull, K=7);
%SGPLOT{distribution=Weibull, K=8);
%SGPLOT(distribution=Lognorma1, K=6);
%SGPLOT(distribution=Lognorma1, K=7);
%SGPLOT{distribution=Lognorma1, K=8);
%SGPLOT(distribution=Norma1, K=6);
%SGPLOT(distribution=Norma1, K=7);
%SGPLOT(distribution=Norma1, K=8);
%SGPLOT{distribution=Uniform, K=6);
%SGPLOT{distribution=Uniform, K=7);
%SGPLOT(distribution=Uniform, K=8);
ods rtf file = "&folder\&dist Median=&MED K=&K..rtf" bodytitle;
%pr±nt(distribution=Weibull, K=6);
%pr±nt(distribution=Weibull, K=7);
%pr±nt(distribution=Weibull, K=8);
%pr±nt(distribution=Lognormal, K=6);
%pr±nt(distribution=Lognormal, K=7);
%pr±nt(distribution=Lognormal, K=8);
%pr±nt(distribution=Normal, K=6);
%pr±nt(distribution=Normal, K=7);
%pr±nt(distribution=Normal, K=8);
%pr±nt(distribution=Uniform, K=6);
%pr±nt(distribution=Uniform, K=7);
%pr±nt(distribution=Uniform, K=8);
ods rtf close;
-------
APPENDIX D
I'iihlc l)-l. Results ill" powoi- ;m;il\sis w lion (lie lowest ;iccc|)(;il)k' I'iilio I.CI. /mCl'T = 0.(»
(l.oiiii(M in;il (lislril)iili(in)
p5MR
(llOll I'S)
S;ini|)k* size
10 II 12 13 14 15 l(> 17 IS IV 20
2
0.2
0.045
0.223
0.126
0.344
0.239
0.257
0.355
0.362
0.440
0.467
0.353
0.4
0.236
0.580
0.476
0.776
0.666
0.678
0.815
0.812
0.900
0.903
0.869
0.5
0.449
0.820
0.770
0.937
0.902
0.899
0.958
0.962
0.982
0.985
0.984
0.6
0.768
0.969
0.955
0.979
0.988
0.980
0.980
0.991
0.963
0.981
0.988
0.7
0.964
0.971
0.985
0.933
0.961
0.931
0.926
0.948
0.886
0.915
0.940
0.8
0.894
0.867
0.914
0.801
0.851
0.761
0.797
0.839
0.743
0.792
0.827
4
0.2
0.048
0.169
0.103
0.259
0.188
0.162
0.295
0.254
0.381
0.343
0.279
0.4
0.176
0.472
0.411
0.682
0.595
0.578
0.749
0.735
0.851
0.844
0.796
0.5
0.367
0.729
0.662
0.895
0.834
0.828
0.924
0.927
0.973
0.972
0.960
0.6
0.638
0.932
0.896
0.984
0.977
0.972
0.989
0.992
0.998
0.999
0.997
0.7
0.919
0.996
0.991
0.999
0.999
0.998
0.997
0.999
0.995
0.998
0.999
0.8
0.994
0.992
0.997
0.976
0.990
0.981
0.971
0.984
0.949
0.971
0.979
6
0.2
0.175
0.207
0.202
0.283
0.240
0.199
0.343
0.266
0.417
0.355
0.304
0.4
0.180
0.474
0.400
0.677
0.600
0.561
0.751
0.706
0.845
0.827
0.794
0.5
0.360
0.703
0.665
0.876
0.826
0.804
0.930
0.916
0.971
0.963
0.956
0.6
0.635
0.917
0.900
0.982
0.976
0.964
0.992
0.993
0.999
0.998
0.999
0.7
0.922
0.994
0.993
1.000
0.999
0.999
1.000
1.000
0.999
1.000
1.000
0.8
0.999
0.999
1.000
0.998
0.999
0.998
0.996
0.999
0.993
0.998
0.999
8
0.2
0.408
0.389
0.438
0.449
0.470
0.371
0.535
0.418
0.594
0.501
0.487
0.4
0.378
0.567
0.551
0.739
0.697
0.635
0.813
0.766
0.886
0.864
0.842
0.5
0.469
0.742
0.731
0.898
0.868
0.831
0.942
0.924
0.979
0.967
0.963
0.6
0.680
0.923
0.919
0.987
0.983
0.966
0.994
0.992
0.998
0.999
0.998
0.7
0.929
0.994
0.994
1.000
1.000
0.999
1.000
1.000
1.000
1.000
1.000
0.8
0.999
1.000
1.000
1.000
1.000
0.999
1.000
1.000
0.999
1.000
1.000
-------
T;il)k' l)-2: Results of power sis w lion I lie lowest ;ieeepl;il)le riilio ')5% I.CI. /mCl'T = 11.7
(l.(iiiii(M in;il (listrihu 1 ion)
Merihm
Siiniple si/o
(hours)
2
1" >1 K
0 ""
JO
I) DOS
11
0.048
_l 2
0.034
_I3
0.110
_I4
0 o"o
15
I) (Hi i
jr.
0.11"
J 7
o.loo
JX
or:
J«>
u.1^9
_2lt
o.loo
0.4
0.072
0.285
0.197
0.465
0.355
0.390
0.489
0.530
0.608
0.619
0.519
0.5
0.197
0.529
0.419
0.697
0.591
0.641
0.699
0.750
0.783
0.811
0.720
0.6
0.440
0.749
0.648
0.856
0.775
0.816
0.844
0.893
0.877
0.907
0.874
0.7
0.677
0.878
0.831
0.896
0.895
0.870
0.886
0.933
0.855
0.909
0.926
0.8
0.800
0.866
0.905
0.796
0.849
0.744
0.795
0.843
0.737
0.786
0.837
4
0.2
0.024
0.056
0.037
0.110
0.064
0.058
0.107
0.093
0.160
0.139
0.090
0.4
0.057
0.227
0.173
0.411
0.312
0.297
0.462
0.423
0.593
0.541
0.469
0.5
0.157
0.430
0.368
0.627
0.561
0.521
0.714
0.669
0.819
0.765
0.745
0.6
0.353
0.666
0.643
0.838
0.827
0.769
0.909
0.890
0.958
0.928
0.935
0.7
0.692
0.894
0.906
0.966
0.970
0.945
0.988
0.983
0.992
0.993
0.996
0.8
0.954
0.986
0.993
0.977
0.991
0.979
0.970
0.985
0.945
0.965
0.982
6
0.2
0.161
0.128
0.142
0.153
0.149
0.099
0.165
0.118
0.200
0.163
0.121
0.4
0.068
0.210
0.157
0.368
0.279
0.258
0.402
0.361
0.542
0.487
0.403
0.5
0.127
0.379
0.300
0.591
0.487
0.466
0.646
0.614
0.775
0.727
0.669
0.6
0.284
0.623
0.559
0.809
0.773
0.735
0.880
0.867
0.941
0.912
0.914
0.7
0.608
0.879
0.873
0.964
0.959
0.944
0.988
0.981
0.997
0.995
0.997
0.8
0.941
0.993
0.996
0.996
0.999
0.998
0.998
0.999
0.995
0.997
0.999
8
0.2
0.394
0.311
0.381
0.331
0.382
0.274
0.390
0.296
0.407
0.320
0.333
0.4
0.273
0.317
0.326
0.461
0.415
0.331
0.525
0.432
0.628
0.540
0.500
0.5
0.260
0.439
0.406
0.621
0.577
0.504
0.712
0.650
0.824
0.771
0.731
0.6
0.362
0.663
0.617
0.844
0.822
0.775
0.904
0.891
0.959
0.941
0.933
0.7
0.653
0.905
0.898
0.979
0.975
0.972
0.993
0.992
0.999
0.999
0.998
0.8
0.959
0.999
0.998
0.999
1.000
1.000
1.000
1.000
0.999
1.000
1.000
-------
1 :il)k' l)-3 Kcsulls or power ;iii;il>sis when 1 lie lowest iicccpliihle r.ilio «)5V
(l.o^noniiiil (lislrihiilioii)
i.( i.
/inCPI' =
II.X
Mcdiiin
P5 MR
S;imnk- si/o
(lion rs)
J"
11
_I2
_I3
_I4
_I5
_K»
_I7
_IX
_l«>
_20
o ""
I) DOS
o n"
I) (I'M)
U.U43
U.U31
mil1)
0 <>s<>
o u^""
-------
Tiihlc l)-4: Results of power ;m;il\sis w lien (lie lowest ;icco|X;il>lo mlio ')5"ii I.CI. /mCl'T = 0.0
(\orm;il (lislrihulioii)
Merihin
Siimnle size
(hours)
2
1" >1K
0 ""
J"
o r,«>
11
U.454
_I2
0
_l 3
0
_I4
0 SS(i
15
0 S4<>
_l(»
o
_I7
u.-ul
_IX
u.s:4
_l«)
0 X"1!)
_2<1
0 0
0.4
0.326
0.711
0.653
0.877
0.825
0.818
0.916
0.912
0.960
0.962
0.952
0.5
0.506
0.858
0.833
0.946
0.934
0.922
0.966
0.974
0.967
0.981
0.985
0.6
0.730
0.951
0.949
0.963
0.978
0.962
0.958
0.979
0.934
0.957
0.969
0.7
0.925
0.957
0.976
0.908
0.940
0.901
0.900
0.923
0.842
0.882
0.915
0.8
0.875
0.846
0.899
0.768
0.821
0.723
0.760
0.812
0.695
0.753
0.791
4
0.2
0.098
0.360
0.291
0.569
0.476
0.448
0.646
0.618
0.775
0.757
0.692
0.4
0.253
0.635
0.562
0.838
0.764
0.749
0.880
0.874
0.950
0.949
0.926
0.5
0.415
0.800
0.747
0.938
0.901
0.899
0.965
0.960
0.989
0.990
0.982
0.6
0.638
0.936
0.913
0.991
0.983
0.975
0.992
0.994
0.996
0.998
0.999
0.7
0.888
0.993
0.987
0.996
0.998
0.995
0.994
0.998
0.988
0.996
0.997
0.8
0.991
0.988
0.995
0.967
0.983
0.973
0.959
0.980
0.934
0.958
0.970
6
0.2
0.088
0.344
0.272
0.552
0.461
0.426
0.648
0.594
0.777
0.743
0.693
0.4
0.246
0.607
0.558
0.820
0.761
0.719
0.889
0.864
0.956
0.936
0.932
0.5
0.408
0.780
0.745
0.930
0.905
0.876
0.966
0.955
0.990
0.986
0.984
0.6
0.638
0.927
0.918
0.987
0.982
0.973
0.996
0.995
1.000
0.998
1.000
0.7
0.893
0.993
0.992
1.000
0.999
0.998
1.000
1.000
0.999
1.000
1.000
0.8
0.997
0.999
0.999
0.996
0.998
0.997
0.994
0.998
0.988
0.996
0.997
8
0.2
0.231
0.395
0.362
0.574
0.504
0.459
0.667
0.605
0.780
0.756
0.705
0.4
0.303
0.618
0.578
0.821
0.765
0.723
0.891
0.860
0.959
0.941
0.934
0.5
0.422
0.785
0.759
0.933
0.910
0.879
0.967
0.956
0.991
0.986
0.987
0.6
0.646
0.927
0.920
0.989
0.984
0.972
0.997
0.994
1.000
0.999
0.999
0.7
0.896
0.992
0.991
1.000
1.000
0.998
1.000
1.000
1.000
1.000
1.000
0.8
0.997
1.000
1.000
1.000
1.000
0.999
0.999
1.000
0.998
0.999
1.000
-------
Tahle l)-5: Results of |)iini'i' sis when llu' lowest aeeeplahle ratio <)5"/n 1.(1. /inCPT =0.7
(Normal (lislril)iilion)
Median p5MR
(lion rs)
Sample si/e
10 II 12 13 14 15 l(> I"7 IS |«) 20
2
0.2
0.044
0.216
0.143
0.390
0.287
0.293
0.429
0.441
0.566
0.559
0.450
0.4
0.136
0.465
0.347
0.662
0.549
0.592
0.684
0.727
0.790
0.813
0.728
0.5
0.262
0.627
0.521
0.803
0.710
0.744
0.801
0.857
0.857
0.892
0.840
0.6
0.460
0.796
0.712
0.891
0.836
0.854
0.877
0.925
0.883
0.920
0.916
0.7
0.690
0.885
0.860
0.882
0.895
0.855
0.867
0.915
0.822
0.879
0.908
0.8
0.795
0.843
0.886
0.761
0.819
0.706
0.759
0.811
0.689
0.743
0.801
4
0.2
0.033
0.175
0.130
0.340
0.257
0.216
0.405
0.346
0.535
0.469
0.413
0.4
0.111
0.369
0.317
0.581
0.516
0.466
0.683
0.622
0.811
0.746
0.709
0.5
0.214
0.526
0.475
0.742
0.704
0.643
0.833
0.797
0.915
0.871
0.870
0.6
0.391
0.718
0.697
0.882
0.881
0.821
0.944
0.925
0.976
0.957
0.964
0.7
0.683
0.905
0.915
0.971
0.979
0.957
0.990
0.988
0.990
0.993
0.996
0.8
0.940
0.983
0.993
0.966
0.984
0.968
0.961
0.978
0.926
0.957
0.973
6
0.2
0.028
0.153
0.105
0.300
0.215
0.191
0.345
0.301
0.483
0.426
0.347
0.4
0.085
0.328
0.259
0.550
0.455
0.416
0.619
0.580
0.754
0.708
0.646
0.5
0.174
0.486
0.416
0.710
0.648
0.600
0.794
0.764
0.886
0.851
0.827
0.6
0.334
0.681
0.635
0.863
0.837
0.798
0.927
0.914
0.970
0.951
0.952
0.7
0.607
0.892
0.885
0.971
0.969
0.958
0.991
0.989
0.997
0.996
0.998
0.8
0.925
0.991
0.995
0.994
0.999
0.996
0.997
0.998
0.992
0.996
0.998
8
0.2
0.185
0.222
0.214
0.351
0.296
0.231
0.416
0.335
0.522
0.448
0.390
0.4
0.162
0.350
0.308
0.560
0.492
0.428
0.653
0.603
0.784
0.730
0.680
0.5
0.205
0.494
0.439
0.726
0.678
0.621
0.809
0.785
0.909
0.875
0.851
0.6
0.358
0.713
0.662
0.890
0.870
0.835
0.945
0.932
0.981
0.971
0.970
0.7
0.642
0.916
0.902
0.985
0.980
0.980
0.996
0.996
0.999
0.999
0.998
0.8
0.943
0.998
0.998
0.999
1.000
0.999
1.000
1.000
0.998
0.999
1.000
-------
Tahle Results of |)iini'i' sis when llu' lowest aeeeplahle ratio *)5% 1.(1. /inCPT = O.X
(Noi'iiiiil (lisli'il)iilion)
Median p5MR
(lion rs)
Sample si/e
10 II 12 13 14 15 l(> I"7 IS |«) 20
2
0.2
0.013
0.069
0.049
0.142
0.109
0.081
0.176
0.136
0.250
0.193
0.154
0.4
0.032
0.142
0.120
0.259
0.214
0.167
0.311
0.259
0.401
0.352
0.305
0.5
0.067
0.210
0.183
0.355
0.314
0.245
0.426
0.382
0.504
0.469
0.433
0.6
0.116
0.328
0.291
0.502
0.464
0.387
0.578
0.559
0.661
0.650
0.628
0.7
0.239
0.537
0.508
0.683
0.685
0.601
0.741
0.776
0.760
0.799
0.824
0.8
0.512
0.768
0.801
0.745
0.802
0.685
0.756
0.807
0.689
0.741
0.800
4
0.2
0.008
0.051
0.035
0.112
0.084
0.055
0.140
0.098
0.208
0.138
0.123
0.4
0.029
0.114
0.093
0.224
0.192
0.133
0.304
0.214
0.420
0.293
0.295
0.5
0.055
0.182
0.169
0.320
0.305
0.221
0.451
0.323
0.572
0.432
0.443
0.6
0.118
0.289
0.310
0.479
0.484
0.363
0.633
0.503
0.750
0.620
0.632
0.7
0.271
0.512
0.528
0.695
0.720
0.616
0.827
0.744
0.892
0.824
0.853
0.8
0.618
0.800
0.844
0.892
0.923
0.854
0.927
0.925
0.911
0.920
0.952
6
0.2
0.005
0.051
0.031
0.115
0.070
0.054
0.129
0.104
0.196
0.155
0.105
0.4
0.022
0.120
0.088
0.247
0.179
0.159
0.297
0.247
0.410
0.347
0.287
0.5
0.044
0.206
0.163
0.377
0.297
0.265
0.453
0.389
0.584
0.509
0.457
0.6
0.111
0.357
0.307
0.558
0.502
0.444
0.664
0.600
0.799
0.721
0.691
0.7
0.285
0.605
0.573
0.796
0.786
0.716
0.885
0.861
0.949
0.907
0.921
0.8
0.680
0.898
0.913
0.968
0.977
0.955
0.990
0.986
0.990
0.993
0.996
8
0.2
0.164
0.131
0.143
0.164
0.163
0.101
0.190
0.135
0.245
0.177
0.146
0.4
0.097
0.152
0.128
0.260
0.204
0.158
0.298
0.248
0.411
0.353
0.283
0.5
0.082
0.213
0.171
0.389
0.303
0.265
0.440
0.401
0.589
0.536
0.445
0.6
0.119
0.362
0.290
0.591
0.496
0.465
0.660
0.635
0.789
0.759
0.695
0.7
0.273
0.638
0.564
0.827
0.773
0.760
0.887
0.891
0.945
0.933
0.918
0.8
0.658
0.923
0.909
0.983
0.976
0.978
0.995
0.995
0.998
0.998
0.998
-------
Table l)-7; Results of power sis w lien (ho lowest acceptable ratio <)5"/n 1.(1. /mCl'T = 0.(>
(I inform (lislril)iilion)
Median p5MR
(lion rs)
Sample si/e
10 II 12 13 14 15 l(> 17 IS |«) 20
2
0.2
0.059
0.266
0.177
0.434
0.319
0.331
0.471
0.450
0.576
0.586
0.491
0.4
0.242
0.561
0.468
0.727
0.657
0.664
0.792
0.772
0.876
0.870
0.824
0.5
0.429
0.788
0.736
0.923
0.880
0.889
0.955
0.949
0.981
0.982
0.974
0.6
0.987
0.995
0.994
0.995
0.995
0.993
0.990
0.998
0.981
0.990
0.996
0.7
0.996
0.990
0.994
0.970
0.981
0.968
0.960
0.977
0.931
0.961
0.975
0.8
0.900
0.904
0.934
0.851
0.886
0.793
0.849
0.875
0.813
0.851
0.883
4
0.2
0.038
0.183
0.126
0.336
0.253
0.226
0.385
0.340
0.499
0.467
0.397
0.4
0.175
0.446
0.372
0.641
0.577
0.547
0.723
0.675
0.814
0.807
0.743
0.5
0.357
0.681
0.627
0.851
0.796
0.802
0.906
0.883
0.947
0.948
0.917
0.6
0.693
0.919
0.876
0.975
0.956
0.960
0.986
0.984
0.996
0.996
0.991
0.7
1.000
1.000
1.000
1.000
1.000
0.999
0.999
1.000
0.999
0.999
1.000
0.8
0.999
0.997
1.000
0.996
0.996
0.994
0.990
0.998
0.981
0.990
0.996
6
0.2
0.037
0.161
0.114
0.313
0.228
0.198
0.368
0.312
0.485
0.442
0.384
0.4
0.159
0.437
0.359
0.626
0.559
0.541
0.722
0.660
0.819
0.784
0.740
0.5
0.371
0.653
0.620
0.823
0.797
0.766
0.895
0.853
0.946
0.930
0.910
0.6
0.674
0.899
0.879
0.971
0.960
0.951
0.987
0.985
0.996
0.996
0.992
0.7
0.970
0.998
0.999
0.999
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
0.999
0.999
1.000
0.999
0.999
1.000
8
0.2
0.291
0.302
0.323
0.423
0.390
0.299
0.499
0.375
0.564
0.491
0.458
0.4
0.314
0.493
0.463
0.665
0.625
0.561
0.747
0.668
0.834
0.796
0.751
0.5
0.452
0.687
0.668
0.835
0.813
0.784
0.905
0.872
0.949
0.934
0.918
0.6
0.744
0.903
0.902
0.968
0.965
0.952
0.991
0.983
0.995
0.995
0.992
0.7
0.977
0.998
0.997
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
1.000
-------
Tahle l)-K: Results of |)iini'i' sis when llu' lowest aeeeplahle ratio <)5"/n 1.(1. /inCPT =0.7
(I inform (lislrihiilion)
Median |>5MR
(lion rs)
Sample si/e
10 II 12 13 14 15 l(> 17 IS |«) 20
2
0.2
0.016
0.079
0.052
0.168
0.117
0.098
0.184
0.186
0.277
0.267
0.197
0.4
0.073
0.242
0.172
0.417
0.310
0.335
0.445
0.489
0.570
0.601
0.491
0.5
0.180
0.470
0.371
0.645
0.540
0.584
0.662
0.711
0.725
0.780
0.683
0.6
0.543
0.731
0.636
0.803
0.711
0.767
0.783
0.839
0.826
0.874
0.808
0.7
0.624
0.847
0.757
0.890
0.853
0.855
0.885
0.914
0.896
0.931
0.918
0.8
0.820
0.904
0.914
0.845
0.885
0.803
0.841
0.882
0.813
0.856
0.882
4
0.2
0.014
0.066
0.041
0.145
0.100
0.073
0.162
0.145
0.266
0.220
0.155
0.4
0.055
0.203
0.148
0.351
0.285
0.251
0.425
0.379
0.557
0.494
0.432
0.5
0.136
0.365
0.305
0.551
0.499
0.442
0.648
0.594
0.746
0.690
0.655
0.6
0.354
0.600
0.572
0.750
0.761
0.670
0.854
0.807
0.908
0.872
0.874
0.7
0.782
0.850
0.876
0.934
0.947
0.896
0.973
0.954
0.989
0.976
0.980
0.8
1.000
0.999
1.000
0.994
0.995
0.995
0.992
0.995
0.984
0.991
0.996
6
0.2
0.016
0.055
0.035
0.129
0.079
0.062
0.141
0.120
0.224
0.190
0.125
0.4
0.045
0.180
0.118
0.313
0.244
0.209
0.367
0.334
0.493
0.450
0.371
0.5
0.116
0.330
0.252
0.496
0.421
0.384
0.572
0.525
0.672
0.638
0.581
0.6
0.280
0.549
0.493
0.713
0.683
0.616
0.799
0.764
0.875
0.848
0.828
0.7
0.665
0.839
0.846
0.931
0.933
0.896
0.970
0.956
0.992
0.979
0.982
0.8
1.000
1.000
1.000
1.000
1.000
1.000
0.999
1.000
0.999
1.000
1.000
8
0.2
0.268
0.210
0.267
0.257
0.268
0.179
0.297
0.215
0.358
0.267
0.244
0.4
0.203
0.248
0.240
0.367
0.331
0.245
0.434
0.367
0.542
0.467
0.400
0.5
0.203
0.356
0.318
0.513
0.467
0.394
0.615
0.554
0.713
0.660
0.612
0.6
0.338
0.576
0.550
0.741
0.722
0.661
0.838
0.806
0.902
0.877
0.854
0.7
0.693
0.892
0.882
0.959
0.956
0.942
0.982
0.980
0.995
0.990
0.990
0.8
0.996
0.999
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
-------
Tahle Results of|)iini'r sis when (lie lowest acceptable ratio <)5"/n 1.(1. /inCPT = O.X
(I inform (lislril)iili(in)
Median |>5MR
(lion rs)
Sample si/e
10 II 12 13 14 15 l(> 17 IS |«) 20
2
0.2
0.009
0.031
0.023
0.065
0.053
0.031
0.081
0.051
0.120
0.091
0.070
0.4
0.027
0.070
0.057
0.144
0.113
0.084
0.178
0.128
0.253
0.190
0.162
0.5
0.047
0.121
0.119
0.227
0.192
0.142
0.275
0.211
0.344
0.278
0.246
0.6
0.131
0.205
0.200
0.313
0.283
0.221
0.375
0.322
0.476
0.409
0.373
0.7
0.142
0.364
0.315
0.530
0.483
0.418
0.609
0.595
0.703
0.692
0.651
0.8
0.641
0.864
0.871
0.837
0.879
0.795
0.841
0.880
0.813
0.856
0.882
4
0.2
0.003
0.017
0.009
0.048
0.024
0.019
0.050
0.038
0.081
0.063
0.033
0.4
0.013
0.061
0.038
0.121
0.090
0.066
0.138
0.103
0.205
0.160
0.122
0.5
0.031
0.112
0.080
0.195
0.159
0.114
0.249
0.185
0.342
0.251
0.231
0.6
0.092
0.198
0.194
0.309
0.325
0.211
0.428
0.317
0.528
0.403
0.428
0.7
0.303
0.374
0.415
0.523
0.564
0.428
0.673
0.562
0.755
0.654
0.675
0.8
0.766
0.735
0.806
0.805
0.860
0.769
0.902
0.839
0.916
0.874
0.899
6
0.2
0.006
0.015
0.009
0.041
0.019
0.017
0.042
0.029
0.069
0.054
0.029
0.4
0.010
0.049
0.028
0.110
0.069
0.050
0.120
0.101
0.189
0.160
0.106
0.5
0.026
0.101
0.065
0.197
0.142
0.114
0.214
0.201
0.327
0.284
0.219
0.6
0.074
0.238
0.164
0.368
0.298
0.271
0.421
0.382
0.541
0.480
0.426
0.7
0.246
0.461
0.422
0.608
0.611
0.521
0.732
0.656
0.808
0.751
0.748
0.8
0.773
0.824
0.860
0.913
0.939
0.880
0.966
0.940
0.986
0.970
0.978
8
0.2
0.261
0.178
0.247
0.181
0.220
0.141
0.215
0.139
0.214
0.146
0.159
0.4
0.175
0.129
0.162
0.175
0.176
0.100
0.203
0.141
0.245
0.184
0.143
0.5
0.134
0.144
0.135
0.214
0.185
0.128
0.253
0.206
0.338
0.282
0.223
0.6
0.119
0.224
0.182
0.354
0.289
0.254
0.418
0.378
0.529
0.492
0.407
0.7
0.223
0.485
0.419
0.652
0.594
0.564
0.730
0.701
0.803
0.794
0.748
0.8
0.679
0.868
0.851
0.939
0.942
0.916
0.976
0.968
0.991
0.984
0.987
-------
Figure D-l: Power curves of study design when the lowest acceptable ratio 95% LCLmCPi/mCPT = 0.6 (Lognormal distributions).
Lognormal median = 2 hours, K = 0.6
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Lognormal median = 4 hours, K = 0.6
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Lognormal median = 6 hours, K = 0.6
Lognormal median = 8 hours, K = 0.6
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
-------
Figure D-2: Power curves of study design when the lowest acceptable ratio 95% LCLmCPi/mCPT = 0.7 (Lognormal distributions).
Lognormal median = 2 hours, K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Lognormal median = 4 hours, K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Lognormal median = 6 hours, K = 0.7
Lognormal median = 8 hours, K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
-------
Figure D-3: Power curves of study design when the lowest acceptable ratio 95% LCLmCPi/mCPT = 0.8 (Lognormal distributions).
Lognormal median = 2 hours, K = 0.8
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Lognormal median = 4 hours, K = 0.8
-*~
"Mr
/ /AX
/X / v
/ v /v
^
/
/
/
/
' -y,' —-
N 4
14 16
N
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Lognormal median = 6 hours, K = 0.8
Lognormal median = 8 hours, K = 0.8
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
-------
Figure D-4: Power curves of study design when the lowest acceptable ratio 95% LCLmCPi/mCPT = 0.6 (Normal distributions).
Normal median = 2 hours, K = 0.6
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Normal median = 4 hours, K = 0.6
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Normal median = 6 hours, K = 0.6
Normal median = 8 hours, K = 0.6
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
-------
Figure D-5: Power curves of study design when the lowest acceptable ratio 95% LCLmCPi/mCPT = 0.7 (Normal distributions).
Normal median = 2 hours, K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 0.6 |
Normal median = 4 hours, K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 A 0.6 ~ 0./ 0.8 |
Normal median = 6 hours, K = 0.7
Normal median = 8 hours, K = 0.7
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
-------
Figure D-6: Power curves of study design when the lowest acceptable ratio 95% LCLmCPi/mCPT = 0.8 (Normal distributions).
Normal median = 2 hours, K = 0.8
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 i 0.6 |
Normal median = 4 hours, K = 0.8
15%-tile/median ratio o 0.2 + 0.4 x 0.5 A 0.6 ~ 0./ 0.8 |
Normal median = 6 hours, K = 0.8
Normal median = 8 hours, K = 0.8
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
-------
Figure D-7: Power curves of study design when the lowest acceptable ratio 95%
LCLmCPT/mCPT = 0.6 (Uniform distributions).
Uniform median = 2 hours, K = 0.6
N
15%-tHe/median ratio o 0.2 + 0.4 x 0.5 a 0.6 0.7 Q.8~|
Uniform median = 4 hours, K = 0.6
N
15%-tHe/median ratio o 0.2 + 0.4 x 0.5 a 0.6 0.7 0.S |
-------
Uniform median = 6 hours, K = 0.6
N
|5%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
Uniform median = 8 hours, K = 0.6
N
|5%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
-------
Figure D-8: Power curves of study design when the lowest acceptable ratio 95%
LCLmCPT/mCPT = 0.7 (Uniform distributions).
Uniform median = 2 hours, K = 0.7
1.0-
10 12 14 16 18 20
N
15%-tHe/median ratio o 0.2 + 0.4 x Q.5 ^0.6 Q.7 0.8 |
Uniform median = 4 hours, K = 0.7
N
|5%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 * 0.8 |
-------
Uniform median = 6 hours, K = 0.7
N
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 *• 0.8 |
Uniform median = 8 hours, K = 0.7
N
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 -t 0.8 |
-------
Figure D-9: Power curves of study design when the lowest acceptable ratio 95%
LCLmCPT/mCPT = 0.8 (Uniform distributions).
Uniform median = 2 hours, K = 0.8
1.0-
10 12 14 16 18 20
N
15%-tHe/median ratio o 0.2 + 0.4 x 0.5 a 0.6 0./ Q.8~|
Uniform median = 4 hours, K = 0.8
1.0-
10 12 14 16 18 20
N
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 P./ 0.8 |
-------
Uniform median = 6 hours, K = 0.8
n 1 1 1 1 r
10 12 14 16 18 20
N
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 *• 0.8 |
Uniform median = 8 hours, K = 0.8
N
15%-tile/median ratio o 0.2 + 0.4 x 0.5 a 0.6 ~ 0.7 -t 0.8 |
-------
ATTACHMENT 5
EPA Standard Rate of Application
Test products should be applied at 1 g/600 cm2 for aerosols and lotions/wipes. For pump sprays
the Agency recommends applying 0.5 g product/600 cm2. These application rates are based on
dosimetry testing conducted in previous repellent efficacy studies since 2006, which have been
reviewed by the EPA and HSRB.
In April 2015, the HSRB reviewed a protocol conducted by SC Johnson and agreed to the use of
the EPA's standard application rate to replace dosimetry testing.21
Table 1. Combined results of dosimetry testing from skin-applied repellent studies reviewed by
EPA and HSRB since 2006 for three formulation types.
Formulation
Type
Total No. of Subjects Across
Dosimetry Phase Studies
Reviewed
Mean
Dose (g/600 cm2)
+ 1 SD
Dose range
(g/600 cm2 )
Lotion/wipe
112
0.933 +0.299
0.63-1.23
Pump spray
92
0.434 + 0.113
0.32-0.55
Aerosol
25
0.815 + 0.262
0.55-1.08
21 Dawson. Liza. April 22-23. 2015 EPA Human Studies Review Board Meeting Report.
https://www.epa.gov/sites/production/files/2015-06/documents/hsrb april 2015 meeting final report.pdf. p. 12.
------- |