This community innovation has been accepted at the 2025 DHIS2 Annual Conference
Piloting the CHAP Platform in Mozambique
Recognizing the significant impact of climate change on malaria transmission, the Mozambique malaria program sought to enhance disease control by adopting an Early Warning System (EWS). This system aims to proactively identify patterns that precede outbreaks, including those related to climate variability. A key challenge was effectively integrating climate data into the EWS for accurate outbreak predictions. To address this, the program transitioned to the Climate Health Adaptation Platform (CHAP). CHAP is a platform that leverages machine learning models to forecast climate sensitive health outcomes, including malaria. By adopting CHAP, we expected to: i) facilitate seamless integration of climate data into the forecasting process, enhancing the accuracy of outbreak predictions; ii) provide a standardized framework for developing and deploying EWS, streamlining system architecture and maintenance; iii) allow for the inclusion of customized models, not only for malaria but also for other climate sensitive diseases, making it a versatile platform for national health surveillance. While the initial vision of integrating CHAP with multiple DHIS2 instances was modified due to technical limitations, the current implementation still leverages DHIS2 for data integration and dissemination of forecasts. In addition, by integrating CHAP into its operations, the malaria program expects to: a) improve the accuracy of malaria case forecasts utilizing both local and globally developed models within the CHAP platform; b) enable proactive interventions integrating forecasts into DHIS2, allowing stakeholders to proactively implement prevention and control measures, such as vaccination campaigns and public health warnings, before outbreaks occur. The adoption of CHAP represents a significant step forward in the Mozambique malaria program’s efforts to enhance disease surveillance and control.
Primary Author: David Mondlane
Keywords:
Machine Learning, AI, Models, DHIS2, Climate and Health, CHAP
2 Likes