Background: There is a growing interest in the scientific community, funding agencies and Ministries of Health (MoH) to control and eventually eliminate neglected tropical diseases (NTDs). Helminth (parasitic worm) infections, particularly soil-transmitted helminthiasis, schistosomiasis and food-borne trematodiasis, are the most common NTDs. Indeed, half of the world’s population is at risk of infection with one or several of these helminths and more than a billion people are currently infected. While considerable progress has been made over the past 10-15 years in Asia and the Americas regarding the control of soil-transmitted helminthiasis and schistosomiasis, little has been achieved in Africa. Clearly, there is a need for up-to-date and reliable maps of the geographical distribution of the NTDs and the number of infected individuals, including underlying risk factors, so that interventions can be targeted in a spatially explicit and cost-effective manner. The World Health Organization recommends regular administration of anthelminthic drugs to at-risk populations, and hence maps are required to show the areas where interventions are necessary. While large-scale mapping of schistosomiasis has recently come to fruition, with regard to soil-transmitted helminthiasis, apart from a few local studies, continental high-spatial resolution maps and assessment of the diseases determinants taking into account their complex interrelations to demography, ecology and socio-economy are not available. Empirical maps of disease distribution over large areas are based on compiled historical data. These data are not standardised in terms of age groups surveyed, survey periods, diagnostic techniques employed, disease outcome measures (e.g. morbidity questionnaire vs. direct parasitological tests) and levels of aggregation (survey locations or regional summaries). Additionally, the diseases-environment relation is not uniform over the study region. Bayesian geostatistical models are the state-of-the-art methodology for predicting disease distribution at high spatial scales. However, application of these models is not practical for data collected over large number of locations due to inherent computational challenges.
Goal and specific objectives: The goal of this Research Module, which is linked to a Training Module entitled “SSPH+ PhD Training Program in Public Health”, is to assess the spatio-temporal dynamics of soil-transmitted helminthiasis across Africa. The project will pursue the following interrelated specific objectives: (i) to assess the contribution of climatic, demographic, environmental and socio-economic determinants on the geographical distribution of soil-transmitted helminthiasis in Africa; (ii) to estimate and obtain continent-wide high-spatial resolution maps of the disease burden (transmission risk, number of infected people and burden estimates); (iii) to assess changes of the disease distribution over time and space; and (iv) to generate continent-wide high-spatial resolution maps of soil-transmitted helminthiasis and schistosomiasis co-endemicity. Methods of investigation: We propose to accomplish these objectives by employing and further developing state-of-the-art Bayesian (a) binomial models for very large non-stationary geostatistical data; (b) variable selection approaches in a geostatistical regression model; (c) mathematical transmission models to standardize age-heterogeneous historical survey data; and (d) shared component geostatistical models for modelling co-endemicity from single-disease independent surveys. Zero-inflated model specifications will be considered for sparse data. The models will be fitted (i) using Markov chain Monte Carlo (MCMC) and reversible jump (RJMCMC) simulation algorithms; and (ii) analyzing spatially structured diseases data obtained from surveys that took place over the past 30 years in Africa, extracted from peer-reviewed literature, MoH reports, national control programmes that are currently available in an open-access global NTD database (GNTD), which was compiled and being updated by the Swiss TPH and partners.
Proposed time frame: October 2011 to September 2014. Significance: This research will, first, generate up-to-date burden estimates and maps of soil-transmitted helminthiasis and related co-infections at high spatial resolution across Africa. These maps will be of value in gauging needs of control programmes, and as a benchmark for estimating the effectiveness of national control programmes. Second, the research will deepen our understanding of the climatic, demographic, environmental and socio-economic factors and their relative contribution to the transmission of soil-transmitted helminthiasis. Such information is valuable for health planning and resource allocation, not only for morbidity control by means of large-scale administration of drugs, but also reducing exposure and hence controlling transission. Third, the spatio-temporal dynamics of the distribution of soil-transmitted helminthiasis will be determined, including underlying risk factors. Fourth, rigorous data-driven statistical methodologies for modeling the spatio-temporal dynamics of soil-transmitted helminthiasis will be further enhanced, and hence risk factor analysis, mapping and prediction of the soil-transmitted helminthiasis is strengthened, which is relevant for other NTDs. Finally, the GNTD will be expanded by cataloguing historical data pertaining to NTDs in Africa, which is relevant for researchers and disease control managers.