House Screening - Biological and Environmental Control of Disease Vectors - page 128

blocked when screens are installed over the larger
wall apertures.
Celli (1901) conducted one of the fi rst
studies of screening on doors and windows
against mosquitoes. He modifi ed railway
housing along fi ve train lines in Italy - windows
were covered with thin muslin and outer doors
with metallic mesh. In addition, doors were fi xed
to close automatically and an extra door was
installed at the top of the stairs to protect the
bedrooms. In these houses Celli observed that
only 5% (10/207) of the occupying railway
workers and their families acquired malaria.
Presumably the dominant vector, An . labranchiae ,
a highly endophilic and anthropophagic
mosquito, could not fi nd alternative routes of
entry into the houses. In contrast, in unscreened
control houses, mosquitoes were found to enter
continuously and 89% (204/229) of residents
were al icted with malaria.
The importance of doors and windows as
entry points varies between mosquito species.
Anopheles vestitipennis in southern Belize is a
nocturnal endophagic mosquito species. It lands
low on exterior walls before moving in short
fl ights horizontally or vertically. As a result,
doors (44%) and windows (26%) are the primary
routes of entry into huts (Grieco et al ., 2000).
Anopheles pharoensis is another low-fl ying
mosquito that does not enter houses through
eave gaps but uses openings at ground level
(Snow, 1987). In contrast, the screening of
doors or windows alone did not reduce the
number of An . gambiae s.l. caught in experi-
mental huts and local houses in south-east
Tanzania, and only marginally reduced the
densities of Mansonia spp., Cx . quinquefasciatus ,
Cx . theileri and Cx . univittattus compared to
unscreened houses (Ogoma et al ., 2010).

(e.g. for ceilings made of wood or palm fronds),
insecticide-treated panels could be attached to it.
For example Atieli et al . (2009) installed papyrus
ceilings with an integral permethrin-treated
netting panel into typical village houses in
western Kenya, resulting in 84% reduction in
the odds of An . gambiae s.l. presence compared
with unmodifi ed houses. The use of insecticide-
treated curtains for control of disease
transmission has been reviewed and contrasted
with bed nets (e.g. Beach et al ., 1993; Lengeler,
2004). Yet it could be argued that curtains can
prevent mosquito house entry in a way that nets
do not, and therefore a better comparison could
be made with other house screening techniques.
Curtains draped across internal walls, covering
doorways, windows and even eaves have been
advocated by several researchers for mosquito
(Majori et al ., 1987; Procacci et al ., 1991; Curtis
et al ., 1996; Kachur et al ., 1999) and endophilic
sandfl y control (Majori et al ., 1989; Elnaiem
et al ., 1999; Kroeger et al ., 2002; Moosa-Kazemi
et al ., 2007). Insecticides (mainly pyrethroids)
have been tested on a range of curtain materials
including hessian (Poopathi and Rao, 1995),
cotton, polyester, polyethylene and poly-
propylene (Curtis et al ., 1996). The results
against mosquitoes were encouraging, showing
reductions of 50-99.5% in entomological out-
come measures such as estimated human-biting
rate indoors, indoor mosquito densities or blood-
feeding success. In one community trial, the use
of permethrin-impregnated curtains resulted in
substantial reductions in parasite densities
(down by approximately 50%) and signifi cantly
increased haemoglobin concentration levels (by
0.4g dl −1 ) in children aged 6-59 months (Diallo
et al ., 1999; Habluetzel et al ., 1999) compared to
those in a control group with no curtains. The
results against sandfl ies and leishmaniasis were
also extremely promising; in one randomized
controlled trial, the incidence of cutaneous
leishmaniasis in the intervention group using
lambda-cyhalothrin-treated sheets dropped to
0% but increased in the control arm to 8%, from
the baseline year rate of 5% (Kroeger et al .,
2002). Similarly, the mean number of sandfl ies
indoors dropped from 15 per trap to 2per trap in
the intervention arm but did not change in the
control arm (Kroeger et al ., 2002).
Strikingly, it is apparent that the physical
barrier ef ect against mosquitoes, but not against

7.4.4 Curtains and insecticide use in
house screening

Although the focus of this topic is non-chemical
tools for vector control, it is worth briefl y
mentioning the potential of insecticide use in
house screening. If the screening is made from a
netting material then it can be impregnated with
insecticide following standard procedures for
treating conventional bed nets. Even if the
material itself is not suitable for impregnation