The DHS Program > Publications > Improving Estimates of Insecticide-Treated Mosquito Net Coverage from Household Surveys: Using Geographic Coordinates to Account for Endemicity and Seasonality (English)
Estimates of ITN use can be biased downward in countries with substantial variation in malaria endemicity or when part of the country is surveyed outside malaria season and the other part during. Endemicity data from the Malaria Atlas Project (MAP) 2007 and 2010 raster, as well as the Mapping Malaria Risk in Africa (MARA) suitability for transmission raster, were used to quantify risk of malaria from population-based surveys (DHS, MIS, and an AMP). In Kenya, Namibia, Rwanda, Tanzania, and Zimbabwe; stratification of households by endemicity zones shows that households in the high/intermediate endemicity zones are significantly more likely to own at least one ITN compared with households in the no/low zones or with all households. Stratification of households by seasonality shows that in some countries people living in areas surveyed during the malaria season are significantly more likely to have slept under an ITN the night before the survey compared with the national average. These results remain significant in logistic regressions that control for other factors. The analysis demonstrates that malaria endemicity and seasonality are crucial factors to be taken into account when examining ITN ownership and use. Countries with heterogeneity of malaria transmission could benefit from the approach used in this paper. Seasonality has an influence on ITN use that is little understood and difficult to analyze, especially in cross-sectional population-based surveys. This analysis suggests that accounting for malaria endemicity and seasonality in ITN ownership and use is possible when GPS locations are available from population-based surveys.