Environmental and socio-economic risk modelling for Chagas disease in Bolivia

Mischler, P. and Kearney, M. and McCarroll, J. C. and Scholte, R. G. C. and Vounatsou, P. and Malone, J. B.. (2012) Environmental and socio-economic risk modelling for Chagas disease in Bolivia. Geospatial health, 6 (3). pp. 59-66.

PDF - Published Version
Available under License CC BY-NC (Attribution-NonCommercial).


Official URL: http://edoc.unibas.ch/dok/A6094154

Downloads: Statistics Overview


Accurately defining disease distributions and calculating disease risk is an important step in the control and prevention of diseases. Geographical information systems (GIS) and remote sensing technologies, with maximum entropy (Maxent) ecological niche modelling computer software, were used to create predictive risk maps for Chagas disease in Bolivia. Prevalence rates were calculated from 2007 to 2009 household infection survey data for Bolivia, while environmental data were compiled from the Worldclim database and MODIS satellite imagery. Socio-economic data were obtained from the Bolivian National Institute of Statistics. Disease models identified altitudes at 500-3,500 m above the mean sea level (MSL), low annual precipitation (45-250 mm), and higher diurnal range of temperature (10-19 °C; peak 16 °C) as compatible with the biological requirements of the insect vectors. Socio-economic analyses demonstrated the importance of improved housing materials and water source. Home adobe wall materials and having to fetch drinking water from rivers or wells without pump were found to be highly related to distribution of the disease by the receiver operator characteristic (ROC) area under the curve (AUC) (0.69 AUC, 0.67 AUC and 0.62 AUC, respectively), while areas with hardwood floors demonstrated a direct negative relationship (-0.71 AUC). This study demonstrates that Maxent modelling can be used in disease prevalence and incidence studies to provide governmental agencies with an easily learned, understandable method to define areas as either high, moderate or low risk for the disease. This information may be used in resource planning, targeting and implementation. However, access to high-resolution, sub-municipality socio-economic data (e.g. census tracts) would facilitate elucidation of the relative influence of poverty-related factors on regional disease dynamics.
Faculties and Departments:09 Associated Institutions > Swiss Tropical and Public Health Institute (Swiss TPH) > Department of Epidemiology and Public Health (EPH) > Biostatistics
UniBasel Contributors:Vounatsou, Penelope
Item Type:Article, refereed
Article Subtype:Research Article
Note:Publication type according to Uni Basel Research Database: Journal article
Related URLs:
Identification Number:
edoc DOI:
Last Modified:13 Oct 2017 08:04
Deposited On:24 May 2013 09:20

Repository Staff Only: item control page