Guinea (Graves et al., 1988), and Pakistan (Strickland et al., 1987). Local factors (in addition to IRS and insecticide-impregnated materials) that affect the microepidemiology of malaria are house siting, screening and construction, proximity of animals to human dwellings, and use of mosquito deterrents such as repellants, aerosols, and fumigants.

Despite the fact that anopheline mosquitoes can fly substantial distances, the proximity of houses or villages to a breeding site strongly influences malaria risk, especially where breeding sites are restricted. In a suburb of Dakar, Senegal, malaria prevalence rose steeply from the center to the edge of town adjacent to marshy breeding sites of Anopheles arabiensis (Trape et al., 1992). In Sri Lanka, the risk of malaria was much higher among those who lived in poor quality houses within 2.5 km of a river where A. culicifacies bred (Gunawardena et al., 1998).

House design and construction also influence the risk of malaria (Schofield and White, 1984). Eaves—which allow interior ventilation, and the escape of smoke from cooking fires—are a common feature that facilitate mosquito access to sleeping areas in houses in the tropics. Using mud or plaster to fill in eaves (Lindsay and Snow, 1988), or hanging eaves curtains (Curtis et al., 1992) reduce human-vector contact. In Sri Lanka, Gunawardena et al. (1998) estimated that the cost of upgrading all low-quality housing (whose residents suffered a fourfold risk of malaria compared to families living in well-constructed houses in the same locale) would be balanced by savings in malaria treatment costs over a period of 7 years.

In some communities, animals live in or near houses. Zooprophylaxis is a term that suggests the possible diversion of mosquito bites from humans to nearby animals. However, this diversion depends entirely on the biting habits of the local vector and varies from species to species. In some cases, livestock may actually attract certain mosquitoes that would otherwise avoid human habitats, resulting in increased malaria exposure to household members (Hewitt et al., 1994; Bouma and Rowland, 1995; Mouchet, 1998). In Pakistani and Afghan refugee camps, malaria cases were concentrated in communities that kept cattle, presumably because the local vectors A. culicifacies and A. stephensi were preferentially attracted to these households (Bouma and Rowland, 1995).