Mapping malaria risk in East Africa at high spatial resolution

Kenya Medical Research Institute/Wellcome Trust Research Programme, Nairobi, Kenya.
Tropical Medicine & International Health (Impact Factor: 2.33). 07/2005; 10(6):557-66. DOI: 10.1111/j.1365-3156.2005.01424.x
Source: PubMed


Malaria risk maps have re-emerged as an important tool for appropriately targeting the limited resources available for malaria control. In Sub-Saharan Africa empirically derived maps using standardized criteria are few and this paper considers the development of a model of malaria risk for East Africa.
Statistical techniques were applied to high spatial resolution remotely sensed, human settlement and land-use data to predict the intensity of malaria transmission as defined according to the childhood parasite ratio (PR) in East Africa. Discriminant analysis was used to train environmental and human settlement predictor variables to distinguish between four classes of PR risk shown to relate to disease outcomes in the region.
Independent empirical estimates of the PR were identified from Kenya, Tanzania and Uganda (n = 330). Surrogate markers of climate recorded on-board earth orbiting satellites, population settlement, elevation and water bodies all contributed significantly to the predictive models of malaria transmission intensity in the sub-region. The accuracy of the model was increased by stratifying East Africa into two ecological zones. In addition, the inclusion of urbanization as a predictor of malaria prevalence, whilst reducing formal accuracy statistics, nevertheless improved the consistency of the predictive map with expert opinion malaria maps. The overall accuracy achieved with ecological zone and urban stratification was 62% with surrogates of precipitation and temperature being among the most discriminating predictors of the PR.
It is possible to achieve a high degree of predictive accuracy for Plasmodium falciparum parasite prevalence in East Africa using high-spatial resolution environmental data. However, discrepancies were evident from mapped outputs from the models which were largely due to poor coverage of malaria training data and the comparable spatial resolution of predictor data. These deficiencies will only be addressed by more random, intensive small areas studies of empirical estimates of PR.

Download full-text


Available from: Andrew Tatem,
  • Source
    • "Faso (Baragatti et al. 2009). Omumbo et al. (2005) identified urbanization as a predictor of malaria prevalence . Moffett et al. (2007) identified the growing population density as major factor in determining the malaria risk in Africa. "
    [Show abstract] [Hide abstract]
    ABSTRACT: One of the features of medical geography that has made it so useful in health research is statistical spatial analysis, which enables the quantification and qualification of health events. The main objective of this research was to study the spatial distribution patterns of malaria in Rawalpindi district using spatial statistical techniques to identify the hot spots and the possible risk factor. Spatial statistical analyses were done in ArcGIS, and satellite images for land use classification were processed in ERDAS Imagine. Four hundred and fifty water samples were also collected from the study area to identify the presence or absence of any microbial contamination. The results of this study indicated that malaria incidence varied according to geographical location, with eco-climatic condition and showing significant positive spatial autocorrelation. Hotspots or location of clusters were identified using Getis-Ord Gi* statistic. Significant clustering of malaria incidence occurred in rural central part of the study area including Gujar Khan, Kaller Syedan, and some part of Kahuta and Rawalpindi Tehsil. Ordinary least square (OLS) regression analysis was conducted to analyze the relationship of risk factors with the disease cases. Relationship of different land cover with the disease cases indicated that malaria was more related with agriculture, low vegetation, and water class. Temporal variation of malaria cases showed significant positive association with the meteorological variables including average monthly rainfall and temperature. The results of the study further suggested that water supply and sewage system and solid waste collection system needs a serious attention to prevent any outbreak in the study area.
    Environmental Monitoring and Assessment 09/2015; 187(9):1-15. DOI:10.1007/s10661-015-4779-9 · 1.68 Impact Factor
  • Source
    • "Thus, next to environmental (including climatic) factors that influence the spatial distribution of malaria, it is important to also take into consideration the range of socioeconomic, demographic, political, and behavioral factors that impact people’s susceptibility and (lack of) resilience to the disease [17-19]. Several papers have been published on factors that influence the spread and spatial distribution of the disease [21,24], including eastern Africa [25], and there are a few papers assessing malaria risk, that, besides environmental factors, also integrate socioeconomic and demographic factors [9,26-29]. To date, however, only few studies have been published on vulnerability to vector-borne diseases [17,18,30-32], and malaria in particular [9]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Despite efforts in eradication and control, malaria remains a global challenge, particularly affecting vulnerable groups. Despite the recession in malaria cases, previously malaria free areas are increasingly confronted with epidemics as a result of changing environmental and socioeconomic conditions. Next to modeling transmission intensities and probabilities, integrated spatial methods targeting the complex interplay of factors that contribute to social vulnerability are required to effectively reduce malaria burden. We propose an integrative method for mapping relative levels of social vulnerability in a spatially explicit manner to support the identification of intervention measures. Methods Based on a literature review, a holistic risk and vulnerability framework has been developed to guide the assessment of social vulnerability to water-related vector-borne diseases (VBDs) in the context of changing environmental and societal conditions. Building on the framework, this paper applies spatially explicit modeling for delineating homogeneous regions of social vulnerability to malaria in eastern Africa, while taking into account expert knowledge for weighting the single vulnerability indicators. To assess the influence of the selected indicators on the final index a local sensitivity analysis is carried out. Results Results indicate that high levels of malaria vulnerability are concentrated in the highlands, where immunity within the population is currently low. Additionally, regions with a lack of access to education and health services aggravate vulnerability. Lower values can be found in regions with relatively low poverty, low population pressure, low conflict density and reduced contributions from the biological susceptibility domain. Overall, the factors characterizing vulnerability vary spatially in the region. The vulnerability index reveals a high level of robustness in regard to the final choice of input datasets, with the exception of the immunity indicator which has a marked impact on the composite vulnerability index. Conclusions We introduce a conceptual framework for modeling risk and vulnerability to VBDs. Drawing on the framework we modeled social vulnerability to malaria in the context of global change using a spatially explicit approach. The results provide decision makers with place-specific options for targeting interventions that aim at reducing the burden of the disease amongst the different vulnerable population groups.
    International Journal of Health Geographics 08/2014; 13(1):29. DOI:10.1186/1476-072X-13-29 · 2.62 Impact Factor
  • Source
    • "While our models using TWI showed high accuracy, further studies comparing the use of MWI and TWI in larval habitat modeling are needed. NDVI has also been associated with the distribution of malaria [46,47], although some studies have found contradicting results [17,48,49]. NDVI is an indirect measure of available moisture, but NDVI values are additionally influenced by vegetation type and phenology. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Predictive models of malaria vector larval habitat locations may provide a basis for understanding the spatial determinants of malaria transmission. Methods We used four landscape variables (topographic wetness index [TWI], soil type, land use-land cover, and distance to stream) and accumulated precipitation to model larval habitat locations in a region of western Kenya through two methods: logistic regression and random forest. Additionally, we used two separate data sets to account for variation in habitat locations across space and over time. Results Larval habitats were more likely to be present in locations with a lower slope to contributing area ratio (i.e. TWI), closer to streams, with agricultural land use relative to nonagricultural land use, and in friable clay/sandy clay loam soil and firm, silty clay/clay soil relative to friable clay soil. The probability of larval habitat presence increased with increasing accumulated precipitation. The random forest models were more accurate than the logistic regression models, especially when accumulated precipitation was included to account for seasonal differences in precipitation. The most accurate models for the two data sets had area under the curve (AUC) values of 0.864 and 0.871, respectively. TWI, distance to the nearest stream, and precipitation had the greatest mean decrease in Gini impurity criteria in these models. Conclusions This study demonstrates the usefulness of random forest models for larval malaria vector habitat modeling. TWI and distance to the nearest stream were the two most important landscape variables in these models. Including accumulated precipitation in our models improved the accuracy of larval habitat location predictions by accounting for seasonal variation in the precipitation. Finally, the sampling strategy employed here for model parameterization could serve as a framework for creating predictive larval habitat models to assist in larval control efforts.
    International Journal of Health Geographics 06/2014; 13(1):17. DOI:10.1186/1476-072X-13-17 · 2.62 Impact Factor
Show more