This abstract has been accepted at the 2024 DHIS2 Annual Conference
session link: AI & Machine Learning (ENG- FR)
Integrating Climatic and Entomological Data with Early Warning Models into DHIS2 for Improved Dengue and Malaria Surveillance in Tanzania
Climate change could affect vector-borne disease (VBD) transmission, geographical spread, and re-emergence. VBDs including malaria and arboviral infections such as dengue pose significant health risks in Tanzania. Effective surveillance systems are crucial for timely prevention and control efforts. Though partially effective vaccines for malaria and dengue exist, disease control is dependent on controlling vector abundance and distribution. Our study aims at developing and integrating a Spatio-temporal Early Warning System (EWS) model for malaria and dengue outbreaks in Tanzania with DHIS2 for analysis and dissemination. An EWS for VBDs is crucial for epidemic preparedness and response by enabling timely action in high-risk areas. The effectiveness of an EWS depends on reliable vector/incidence data, which in combination with climatic data can be used to build forecasting models. Climatic factors (rainfall, temperature, humidity, wind speed) affect mosquito survival and reproduction rates and disease transmission. An EWS integrated into DHIS2 will facilitate data analysis and dissemination to support effective deployment of resources, reducing disease incidence and burden on healthcare and economic systems caused by epidemics. Over a two-year period, mosquitoes from selected households in 14 villages from 7 districts in Tanga region and 4 Shehias from 4 districts in Unguja were sampled monthly in rainy and dry seasons. Advanced analytical techniques like machine learning and statistical modeling are employed to identify relationships between the collected entomological data, disease outbreaks, and climatic derived data, which are used to develop predictive models. These models will help predict high-risk areas and provide timely alerts for targeted interventions. By integrating these models with climatic and entomological data into the DHIS2 platform, Tanzania will establish an enhanced EWS for early detection of dengue and malaria epidemics. This integration will empower health authorities to identify high-risk areas, efficiently allocate resources, and implement targeted interventions to control these disease vectors.
Primary Author: Wilfred Senyoni
Keywords:
DHIS2, Modelling, Vector Borne Disease, Malaria, Dengue