DISCUSSION QUESTIONS FOR MOSQUITO REPELLENT STUDIES

The Human Studies Review Board (HSRB or Board) has discussed and
provided advice to EPA on scientific and ethical issues related to the
conduct of field studies to evaluate the efficacy of mosquito repellent
products.  The HSRB has reviewed both proposals for new field studies
and the results of completed studies.  The HSRB has noted that, although
there are many similarities across studies, not all studies employ the
same study design.  The HSRB has identified several methodological
issues for which additional background information would assist the
Board in its evaluation of such studies.  

BACKGROUND	

Currently, EPA requires all pesticide products that claim to repel
mosquitoes to provide data on the duration of efficacy under field
conditions at two biologically distinct sites.  These data are derived
from human research with subjects who have been treated with the
repellent formulations in the field.  The Agency evaluates the duration
of repellent efficacy for a subject by calculating the time from
application of the repellent to the occurrence of an event indicating an
efficacy failure.  Historically, for field studies of mosquito
repellency, EPA has used the “first confirmed bite” as an indication
of efficacy failure on a test subject.  Several recent studies have
shifted to the “first confirmed landing with intent to bite;” EPA
has accepted this alternative endpoint.  A “confirmed landing” on a
test subject is a mosquito landing followed by a second landing on the
same subject within a specified period of time (usually 30 minutes)
after the initial landing.  

Field studies typically involve 6 – 10 subjects who have been treated
with a defined amount of the test material.  Each subject is then
regularly and repeatedly exposed to ambient mosquito populations for a
fixed interval of time until the subject experiences an efficacy failure
followed by a confirmation with the specified period of time. Mosquito
landing pressure (representing intent to bite) at a site is monitored by
concurrently exposing untreated subjects to mosquito landings.  A study
is considered valid only if there are at least a specified minimum
number of mosquito landings on untreated subjects during each exposure
interval. 

On October 25, 2007, the HSRB will discuss scientific aspects of the
design of field studies to assess the efficacy of mosquito repellents. 
For this meeting the Board has requested consultants to provide
specialized information or assistance to the Board.   The Board is
particularly interested in the frequency, duration and timing of
exposure of subjects to potential mosquito landings.  The Board requests
each consultant to respond briefly to the series of questions below. 
Please send the responses to the HSRB Chair and Designated Federal
Official (DFO) at least one week before the meeting—i.e., by no later
than October 18.  All responses will subsequently be provided to the
other consultants, the HSRB members, and EPA staff for their review, and
will be posted on   HYPERLINK "http://www.regulations.gov" 
www.regulations.gov  under docket ID number, EPA-HQ-ORD-2007-0942.  HSRB
consultants will be available at the meeting to discuss their responses
and address questions from the Board.  The questions for Board
consultant consideration are provided below: 



DISCUSSION QUESTIONS

	What do data show about the variability of the time intervals
between first and subsequent landings in mosquito repellent field
trials?

The time between first and second (or subsequent) landings is likely to
be very variable. Thirty minutes is usually suggested as a long enough
interval to allow for any mosquitoes in the area to land.

	What is the current scientific understanding of how factors other
than repellent efficacy could affect the likelihood that an initial
event—a mosquito landing or mosquito bite—would be “confirmed”
by another similar event within 30 minutes?  Please address at least
these factors:

o	Characteristics of mosquito populations

The assumption is that the susceptibility of the mosquito population
follows a normal distribution, and therefore the first mosquito to land
is likely to be at the edge of that distribution. Statistically, the
variation and abundance of the population will dictate when the second
landing occurs. Presumably, the interval between landings will decrease
as the middle of the distribution curve is reached. 

The literature documents that particular species of mosquitoes are
characteristically more or less susceptible to a given active
ingredient. Therefore, the interval between first and second landing
will depend on the species. This assumes a normal curve response to
repellency, as is observed in the laboratory in dose-response studies.

It is also documented that size of individual mosquitoes affects
avidity, generally with larger mosquitoes being more avid. Therefore, a
population that had good larval nutrition (and more uniform, larger
adults) is probably more likely to have a shorter interval between first
and second landings.

Age of mosquitoes affects avidity, with at least some studies showing
that older mosquitoes can be more avid. If the population is from a
single brood, and therefore of similar age, then two effects will occur
with age. First, the older mosquitoes may be more avid and, second, the
population will decline in numbers as it ages. 

Most species of mosquitoes do not bite when they are gravid or within
the first approximately 12-24 hours after oviposition. Especially if the
current population of adults developed from a single hatching event
(e.g., a rising river hatching a brood of floodwater mosquitoes), there
might be an interval when many are gravid or recently parous and the
effective biting population is low. The low population would cause a
longer interval between first and second landing.

o	Characteristics of test sites

Some conditions are known to discourage mosquito biting and host
seeking. Low temperatures, higher winds, and small phases of the moon
are examples. Presumably, these conditions would lengthen the interval
between first and second landings by influencing the avidity of the
population as a whole.  

o	Characteristics of test subjects

People can be more attractive  to mosquitoes because of inherent
characteristics (skin color, skin chemistry, blood chemistry) or
characteristics associated with activity or temporary conditions
(apocrine sweat, skin temperature, illness, alcohol consumption). People
who are more attractive to biting mosquitoes would experience a shorter
interval between first and second landings if we accept the assumption
that the normal curve describing attraction to that particular host is
narrower.

o	Characteristics of test methods

If more of the person were exposed (e.g., only forearms compared to arms
and legs), then mosquitoes should find the host more quickly  and the
interval between first and second landings would be shorter. This is a
testable hypothesis and I am not sure it has been addressed in the
literature.

If the person is instructed to walk around between tests, or if s/he
moves around even while seated, mosquitoes in the vicinity  will have
stronger visual cues for orientation and the interval between first and
second landings could be shorter. This will be particularly true for
mosquitoes that tend to bite larger animals, like floodwater Aedes,
Anopheles, and some container breeders.

Although we think of most repellents as having very little effect over a
distance, almost all of them actually perform in the olfactory phase.
Therefore, a volatile repellent (e.g., deet or oil of citronella) will
have a greater spatial effect if more volume is used on the individual.
A larger volume of a volatile repellent (i.e., if a larger part of the
body was treated) might make the degradation curve of effectiveness more
shallow and therefore lengthen the time between first and second
landings.

When the concentration of less volatile repellents (e.g., Picaridin,
PMD) decreases, landing can occur without biting. In this case, the same
individual mosquito might land, not bite, then land again. That
situation would result in a very short interval between first and second
landings. 

	Can the impact of such factors on the likelihood or timing of an
initial and confirming event be predicted?  Can it be quantified?   

No. 

At its June 27 - 29, 2007 meeting the Board learned that different
designs with different “length-biased” sampling for mosquito
repellent field studies are in use.  One design exposes subjects to
potential mosquito landings for one minute of every 15 minutes; another
design exposes subjects to potential mosquito landings for five minutes
of every 30 minutes.  The
DFO椠⁳敳慰慲整祬瀠潲楶楤杮愠䌠⁄潣瑮楡楮杮琠敨戠
捡杫潲湵⁤慭整楲污⁳潦⁲桴⁥畊敮㈠‷ₖ㤲‬〲㜰䠠
剓⁂敭瑥湩⹧†桔⁥牰瑯捯汯⁳牡⁥潬摡摥漠⁮桴⁥䑃
‮吠敨敳搠獥杩獮栠癡⁥楤晦牥湥⁴沓湥瑧⵨楢獡摥ₔ
慳灭楬杮‮ഠ

	What is the methodological rationale for the two different designs? 

There is nothing very standard about these timing regimes. 

	Which design is used more widely in the field? Why? 

The 30 minute design (or a one hour design) is much more common in the
literature. One reason is that it is easier to perform. The other reason
is that the perceived precision seems to be adequate given the inherent
imprecision of the measurement. In other words, you are never going to
get consistent protection times on a scale finer than about 30 minutes
and even if you did, would it be meaningful in terms of real world
variability in performance?.

	Can potential effects of variation in the pattern of intermittent
exposure on the results of efficacy testing be isolated from the effects
of other variables?  If so, can the direction or magnitude of the
effects be predicted?  How might these influences be analyzed and
accounted for in collecting, reporting and analyzing repellent efficacy
data?

It might be possible to adjust for wind and temperature, if the studies
have been done for that particular area and those species of mosquitoes.
That way, you would adjust the result for the conditions measured at the
time of the intermittent exposure. A much more serious problem is that
the biting activity of mosquitoes varies systematically with time of
day. It is never really accurate to do a repellent trial continuously
for more than about an hour or two because the avidity of the mosquito
population will start to change significantly. The best designs treat
people the appropriate number of hours before the peak biting time and
the expose all subjects simultaneously. 

Dr. Matt Kramer, a USDA statistician who has served as a consultant, has
suggested that the precision of estimates of Complete Protection Time
(CPT) in repellent testing could be significantly increased by defining
a failure of efficacy as the mean time from treatment to a series of
several landings or bites.  He has stated:

The precision of CPT increases when it is estimated beyond time to
[First Confirmed Bite] FCB or FCLanding.  How well CPT can be estimated
depends on the distribution of so many bites beyond FCB.  The number of
mosquitoes that will bite (n) will determine results of the test.  Each
person in the field should be his/her own control; that way it is
possible to know n per person, and reduce person-to-person variability.

If using the mean time to the first 5 bites, the SE will decrease
proportionally as n increases (n = 5 in this case).  That is equivalent
to an increase in the power of the test of 5 times.  This method allows
for detecting formulation differences near the CPT.

	Does this approach, indeed, increase the precision of estimates of
CPT markedly without requiring additional subjects?

Yes. In an extreme case, it might even decrease exposure by requiring
less replication.

A statistician needs to look at the problem, but I would advocate that
the mean of the times for the 5 bites be used as a single index without
concern about the variability between the 5 mosquitoes. The idea would
be that the average time for the first 5 landings would be the new
statistic used as the CPT.

฀࿾㟾$␸䠀$葝ﻠ葞ﻠ摧§!

☊଀Ɇ㜀$␸䠀$摧ᣝú

栂퀁ༀ킄ᄂ预㟾$␸䠀$葞ː葠ﺘ摧§!
ഀ׆Āː฀킄༂킄㜂$␸䠀$葝ː葞ː摧§!ᄀ	If so, would this
increased precision justify the incremental risk to the subjects
resulting from their exposure to a greater number of mosquito landings?

Yes. If each mosquito is aspirated off before it can bite, using the
first five will not increase the risk at all. The exact time of
aspiration would be required in order to calculate the statistic,
therefore, the volunteer would have to remain exposed for whatever
period was necessary following the first landing. In addition, it is
possible that the increased precision will actually decrease the
exposure of volunteers by reducing the number of replications required.

The increase in precision is probably essential if you are going to have
any hope of comparing studies performed at different times. The best
solution might be to apply Kramer’s suggestion of mean landing of five
mosquitoes to lab tests with standard strains, completely eliminating
risk to the volunteer and greatly increasing the precision of the test.
Field tests have so many variables that comparison of products compared
in separate field trials is hopelessly inaccurate. Field tests might
best be used to build confidence in real performance of a single
product, rather than as comparisons between products. It is important to
remember that the objective of the trials in question is labeling so
that consumers can compare products – not a guarantee or estimate of
protection from vector-borne pathogens. 

Is it practical to test long-lasting repellents to the point of five
landings?

Yes. It is just a matter of applying them long enough before the tests
at 30 minute intervals on an adequate number of volunteers. All
volunteers could be tested simultaneously in order to minimize the
variability of the mosquito population caused by daily variation in
biting avidity. Even in the laboratory, avidity varies during the day. 

There is a need for systematic evaluation of standardized test methods
in order to determine the exact procedures necessary to minimize
variability and maximize precision. Standardization will also help
prevent attempts by manufacturers to seek favorable results through the
use of particular field situations and procedures. 

10/04/07	

	Dr. Daniel Strickman’s Responses to Discussion Questions

 PAGE   

 PAGE   1 

