New Prototype Screened Doors and Windows for Excluding Mosquitoes from Houses: A Pilot Study in Rural Gambia

Abstract

Despite compelling evidence that modern housing protects against malaria, houses in endemic areas are still commonly porous to mosquitoes. The protective efficacy of four prototype screened doors and two windows designs against mosquito house entry, their impact on indoor climate, as well as their use, durability and acceptability was assessed in a Gambian village. A baseline survey collected data on all the houses and discrete household units, each consisting of a front and back room, were selected and randomly allocated to the study arms. Each prototype self-closing screened door and window was installed in six and 12 units, respectively, with six unaltered units serving as controls. All prototype doors reduced the number of house-entering mosquitoes by 59-77% in comparison with the control houses. The indoor climate of houses with screened doors was similar to control houses. Seventy-nine percentage of door openings at night occurred from dusk to midnight, when malaria vectors begin entering houses. Ten weeks after installation the doors and windows were in good condition, although 38% of doors did not fully self-close and latch (snap shut). The new doors and windows were popular with residents. The prototype door with perforated concertinaed screening was the best performing door because it reduced mosquito entry, remained fully functional, and was preferred by the villagers. Screened doors and windows may be useful tools for reducing vector exposure and keeping areas malaria-free after elimination, when investment in routine vector control becomes difficult to maintain.

Figure 1.

Evolution of the louvered door. (A) Typical unscreened Francophone door from West Africa; (B) RooPfs door external view; (C) RooPfs door internal view, showing screening and (D), the concertinaed screened door, D1. This figure appears in color at www.ajtmh.org.

Figure 2.

A stylized plan view of a Gambian line house, showing the separate units, each consisting of two rooms.

Figure 3.

Prototype doors (internal view). Where A is the concertinaed door (D1), B is the blinds door (D2), and C and D are doors with a combination of panels and opaque windows at the top (D3 and D4). This figure appears in color at www.ajtmh.org.

Figure 4.

Details of (A), the concertinaed door (external view) (D1) showing perforations underneath the folded metal, and (B), the blinds door (external view) (D2), where the outer surface is perforated and venetian blinds are on the inner surface. This figure appears in color at www.ajtmh.org.

Figure 5.

Prototype windows (internal view). Where A is the blinds window (W1) and B is a window combining an opaque panel at the top and concertinaed panel at the bottom (W2). This figure appears in color at www.ajtmh.org.

Figure 6.

Entry of (A), nuisance mosquitoes and (B), Anopheles gambiae s.l. into houses with different types of doors. This figure appears in color at www.ajtmh.org.

Figure 7.

Psychrometric chart. (A) Human comfort index from 21.00 hours until 00.00 hours, (B) from 00.00 hours to 06.00 hours. Values in red are the percentage of readings that are comfortable. Black polygon indicates comfort area. From 21.00 until midnight all houses are uncomfortably hot, whereas 00.00 to 06.00 hours they are uncomfortably cold, unless using a sheet or blanket. This figure appears in color at www.ajtmh.org.

Figure 8.

Frequency of door openings during the day (n = 15,027 openings in 30 houses over 4 days). This figure appears in color at www.ajtmh.org.

Figure 9.

Duration of door opening during the day in control and intervention housing units. This figure appears in color at www.ajtmh.org.