Kabore A, Biritwum N-K, Downs PW, Soares Magalhaes RJ, Zhang Y, et al. (2013) Predictive vs. Empiric Assessment of Schistosomiasis: Implications for Treatment Projections in Ghana. PLoS Negl Trop Dis 7(3): e2051. doi:10.1371/journal.pntd.0002051
The challenge of accurately mapping schistosomiasis is a daunting one – particularly because of the highly focal distribution of the disease. Ideally, of course, each specific treatment area would be assessed for infection prevalence and then treated appropriately based on guidelines of the World Health Organization. In practice, however, this is not possible, and a variety of short-cutting techniques have been developed to meet these mapping needs, including geospatial predictive mapping. This paper assesses the accuracy of model-based geostatistics (MBG) predictions for determining treatments projections in Ghana by comparing previously published data using MBG predictions with empirically derived prevalence values for schistosomiasis from school surveys completed at 79 sites. We found that using predictive mapping alone would not have provided reliable information for mass drug administration (MDA) planning – resulting in overtreatment in some areas and most importantly under-treatment in areas that needed it most. Based on our findings, predictive risk mapping cannot be a one-time exercise but must instead be a process that incorporates empiric testing and model refining to create optimised spatial decision support tools that meet the needs of disease control operational managers.
For the Full Article: http://www.plosntds.org/article/info:doi/10.1371/journal.pntd.0002051