The impact of different housing designs on mosquito house entry in The Gambia

Abstract

Executive Summary

Background
Although 80-100% of malaria transmission occurs indoors in sub-Saharan Africa, little is known about how changes to the design of houses affects this risk. Simple house modifications can affect house occupant’s exposure to malaria mosquitoes in rural houses in sub-Saharan Africa. This is important since Africa’s housing stock is changing rapidly from the traditional thatched roofed houses with open eaves to metal-roofed houses with closed eaves. In order to study the effect of different housing features on mosquito house entry I carried out a series of experiments using five single-roomed experimental houses with different housing typologies in Wellingara village, The Gambia. It was also important to consider how the different typologies affected indoor climate since a hot house is likely to reduce the use of long-lasting insecticidal nets indoors. These experiments were conducted to address the following questions;
1. What is the impact of different housing typologies on mosquito house entry and indoor climate?
2. What effect do different gaps around the doors have on mosquito house entry in houses with closed eaves?
3. Are small and large screened windows effective in reducing mosquito house entry in houses with closed eaves and badly-fitting doors?

Methods
The five single-roomed experimental houses were the average size of a single-roomed house in The Gambia built from mud block walls. Three experiments were run. The first experiment tested five different housing typologies: 1) thatched roof, open eaves, badly-fitting unscreened doors; 2) thatched roof, closed eaves, badly-fitting unscreened doors; 3) thatched roof, closed eaves, screened louvered metal doors; 4) metal roof, closed eaves, badly-fitting unscreened doors and 5) a novel ventilated metal roof house, with closed eaves, screened gable windows and screened louvered metal doors. The second experiment investigated the role of gaps around doors in mosquito house entry in houses with closed eaves. The third experiment examined the effect of different sizes and numbers of screened windows in houses with closed eaves and badly-fitting doors. Each experiment ran for five weeks and at the end of each five nights, the building materials were rotated between the houses using a replicated Latin rectangle design. In the first experiment, one healthy adult man slept in each house under a treated bed net (Olyset) for five weeks, whilst in the other experiments two healthy adult men slept in each house each night. In each experiment mosquitoes were collected nightly from each house using a CDC light trap. In the first experiment data loggers were used to measure temperature and relative humidity through the day.

Findings
In the first experiment, closing the eaves in thatch-roofed houses with badly-fitting doors reduced the number of Anopheles gambiae s.l. by 94% (Odds ratio (OR)=0.058, 95% confidence intervals (CI)=0.03-0.11) compared with the traditional thatched-roofed houses with open eaves. A similar reduction in mosquito house entry was also found in thatched-roofed houses with well-fitting screened doors. However, closing the eaves of metal-roofed houses did not reduce the number of mosquitoes entering the house compared to the thatched-roofed house with open eaves. Nonetheless, the number of mosquitoes collected in metal-roofed houses with closed eaves was reduced by 94% (OR=0.057, 95% CI=0.03-0.10) when well-fitting screened doors were added to the building. Similar reductions were observed in other Anopheles spp., Culex spp. and Mansonia spp. There was no significant different in night time mean temperature between thatch roofed, open eaves and the ventilated metal-roofed house with closed eaves, screened doors and windows (thatch roofed, open eaves mean temperature=33.08oC, 95% CI=32.58-33.58oC; ventilated metal-roofed house with closed eaves, screened doors and windows mean temperature=33.81oC, 95% CI=33.25-34.37oC). Whilst it was hotter in metal-roofed houses with closed eaves (mean temperature=34.72oC, 95% CI=34.06-35.39oC) than thatched-roofed houses with open eaves (mean temperature=33.08oC, 95% CI=32.58-33.58oC). In the second experiment, the number of An. gambiae s.l. entering houses was not affected by having a single gap above or below the door compared with houses with gaps at both the top and bottom of the door. However, the number of Culex spp. and Mansonia spp. were reduced if the gap was above the door. In the third experiment, the number of An. gambiae s.l. entering metal-roofed houses with badly fitting doors was reduced as the area of screened window increased. Similar reductions were observed in other Anopheles, Culex spp. and Mansonia spp..

Interpretation
These findings demonstrate that design of a house affects mosquito-house entry. Reductions in An. gambiae s.l. can be achieved by closing the eaves of thatched-roofed houses or adding screened doors to metal-roofed houses with closed eaves. Screened doors and windows will help keep the house cool at night and may lead to great bed net use at hot times of the year. Increasing the number of screened windows to a house will reduce house entry of malaria vectors entering badly-fitting doors. Thus housing features such as closed eaves, screened doors and windows should be included in new buildings.