Disclaimer The following was presented in the form of a progress

Disclaimer
The following was presented in the form of a
progress report and is tentative and incomplete
in nature. Do not use it or quote from it with or
without attribution. Please contact J. Sutcliffe
([email protected]) with questions about this
content or about the current status of the
project.
Mosquito entry through holes
Progress toward a predictive
model of mosquito
bed net entry
J.F. Sutcliffe
Trent University
U.S. Centers for Disease Control
What constitutes a “failed net”?
Currently, absolute measures of bed net damage are all that’s available
to judge net serviceability but evidence shows increasingly that the
relation between damage and the likelihood of mosquito entry is not a
simple one.
Stephen C. Smith, Uday B. Joshi, Mark Grabowsky, Joel Selanikio, Theresa
Nobiya, and Thomas Aapore. (2007). Evaluation of bednets after 38 months of
household use in Northwest Ghana. Am. J. Trop. Med. Hyg., 77(Suppl 6), 243–
248.
By observing mosquitoes around the damaged net, this project aims to
provide a behaviourally-informed model of mosquito net entry risk to use
in assessing bed net serviceability.
“Failed net” project
Behavioural analysis approach to determining a net failure
model uses two lines of investigation
1) How does mosquito ‘pressure’
vary across the bed net?
Experiments with
volunteers in treated
and untreated bed nets
2) How do mosquitoes interact
with holes in the bed net?
Video mosquito hole passage
behaviour in small arenas
Combine results from both approaches
into a model of bed net penetrability
Video observations (approach #2) of responses of
female An. gambiae to holes in netting in small
behavioural arenas reveal:
1) Mosquitoes do not actively orient to holes in netting
2) Hole encounter occurs through a random 2-dimensional search
across the net surface
3) Mosquitoes that encounter holes (fly near or across them), do
not always pass through (enter) holes
- Sutcliffe and Colborn (submitted)
What aspects of hole size and shape determine probabilities of
mosquito-hole encounter and of hole passage?
“Failed net” project
Add video analysis of hole encounter and passage in whole
nets to hole interaction studies (2)
1) How does mosquito ‘pressure’
vary across the bed net?
2) How do mosquitoes interact
with holes in the bed net?
Experiments with
volunteers in treated
and untreated bed nets
Combine results from both approaches
into a model of bed net penetrability
Experimental system
- fibreglass screen untreated “bed net” provides
flat surfaces to video flight close to the net and
to video hole encounter using laser illumination
system
- accommodates modules for mounting holes
- accommodates various sizes/shapes of holes on
net sides and roof
Experimental method
- mount holes, close tent, adjust laser
illumination, enter bed net, release unfed
mosquitoes into tent, turn off lights
- video events at each hole (45min)
- log events on videos using Observer® software
Round (circles) and
rectangular (squares)
holes tested
(lesser of dimensions = width for
rectangular holes)
Laser illumination system
fibreglass screening
in aluminum frames
forms walls and roof
1) green laser
source outside
net
2) red laser
source inside
net
4
3) 30cm X 30cm
sampling area
1) line laser projects along outside of net (1cm away) to illuminate
mosquitoes in two-dimensional search against the net
2) second line laser projects along inside of net (1cm) to illuminate
mosquitoes that have passed through the hole and into the net
3) sampling area around hole
4) camera to record near-net activity in sampling area
Key observations made
Appearance - searching mosquito
illuminated in sampling area
Hole encounter – mosquito that has
appeared crosses hole margin
Net entry – mosquito that has encountered
hole, passes through hole into net
Example of video observations
Right - 12mm X 200mm hole in side of
net (taped to prevent premature
mosquito entry)
Below, left and right –
1) “appearance” in sampling area (two
mosquitoes)
2a) hole “encounter”
2b) net “entry”
2a
1
2b
What hole features predict encounter and entry probabilities?
hole circumference accounts for
almost 75% of per appearance
encounter probability variation*
hole width accounts for almost
83% of the variability in per
encounter entry probability*
R2=0.827
R2=0.746
*data from holes on net sides and ends though roof holes operate similarly
The combined model
Combine encounter and entry relationships in Monte Carlo simulations
of the mean number of mosquito entries per appearance over range of
hole widths (0-100mm) and hole circumferences (0-500mm)
combinations:
Per appearance P of hole passage
3
0.20
0.18
5
4
0.16
0.14
0.12
6
0.10
2
0.08
0.06
0.04
1
0.18-0.20
0.16-0.18
0.14-0.16
0.12-0.14
0.10-0.12
0.08-0.10
0.06-0.08
0.04-0.06
0.02-0.04
0.00-0.02
0.02
0.00
- Relationship allows mean and error estimates of entry probability for any of a range of
hole sizes (e.g. 1-6)
“Failed net” project
How many mosquitoes actually enter the net also depends on whether,
and how many, mosquitoes occur on the parts of the net where holes
are located. To determine this, apply results from mosquito pressure
experiments (1) to modelled hole entry probabilities
1) How does mosquito ‘pressure’
vary across the bed net?
Experiments with
volunteers in treated
and untreated bed nets
2) How do mosquitoes interact
with holes in the bed net?
Video mosquito hole passage
behaviour in small arenas
and in full–sized nets
Combine results from both approaches
into a model of bed net penetrability
Sutcliffe, J. and S. Yin (2014). Behavioural responses of females of two
anopheline mosquito species to human-occupied, insecticide-treated and
untreated bed nets. Malaria Journal 13:294.
Host seeking mosquito pressures against the bed net occur in
species-specific patterns:
e.g. An. gambiae caught on sticky
squares in approx. 1:10:40 ratio
on bed net upper sides and
ends:lower third:roof (Functional
Areas 1-3)
FA3
FA1
FA2
Sample model predictions of mean hourly An. gambiae entry rates (and 95%
confidence limits) using preliminary conversions of pressure values* to appearance
values for three mosquito densities, three hole sizes and three areas of the net
number of mosquitoes present in room**
hole features
width (mm) circum. (mm)
5
17
200
functional
area of net
FA1
FA2
FA3
mean
0.00
0.00
0.00
LCL
0.00
0.00
0.00
20
UCL
0.00
0.00
0.00
mean
0.00
0.00
0.00
LCL
0.00
0.00
0.00
2
UCL
0.00
0.00
0.00
mean
0.00
0.00
0.00
LCL
0.00
0.00
0.00
UCL
0.00
0.00
0.00
15
160
FA1
FA2
FA3
0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00
0.02 0.00 0.07 0.00 0.00 0.01 0.00 0.00 0.00
0.18 0.00 0.72 0.02 0.00 0.07 0.00 0.00 0.01
90
360
FA1
FA2
0.48 0.18 0.78 0.05 0.02 0.08 0.00 0.00 0.01
2.97 1.11 4.82 0.30 0.11 0.48 0.03 0.01 0.05
FA3
32.19 12.07 52.31 3.22 1.21 5.23 0.32 0.12 0.52
*Sutcliffe, J. and S. Yin (2014). Behavioural responses of females of two anopheline mosquito species to human-occupied,
insecticide-treated and untreated bed nets. Malaria Journal 13:294.
**estimates assume this number of mosquitoes is constantly present
Conclusion
Results indicate that it should be possible to estimate means
and error terms for mosquito entry risk into damaged bed
nets using simple measures of hole size (e.g. hole width and
circumference) and information about the “Functional Area”
of the net that holes are located in.
Additional/ongoing work
-
More data collection, especially for roof hole entry rates
Address irregularly-shaped holes
Insecticide treated nets
Insecticide resistant mosquito
Test in field/semi-field conditions