DAC 2023 Theme: DHIS2  & Health Emergencies

Early detection of disease before and during public health emergencies is one of the crucial parts of disease surveillance activities. Early detection entails the surveillance system’s ability to use existing data-gathering processes to inform on the possibility or potential for an outbreak of diseases. Despite existing guidance around threshold setting, there are various challenges to adopting relevant approaches within the existing eIDSR system. Based on the UDSM DHIS2 Lab team’s experience with the implementation of eIDSR in the Tanzania mainland, and the evolving demands for surveillance data fueled by the COVID-19 pandemic and other disease outbreaks, a generification to the development of eIDSR was sought to support the adaption of features available to fit the Zanzibar context, as well as create a foundation for adaption of more other new features when needed again in Tanzania mainland. This was purposely done after studying the initial requirements and observing the need to have a configurable application that would be easily adopted in various surveillance situations. To ensure the adaptability of the solution to other diseases and contexts depending on the needs, the eIDSR has been developed and customized with the ability to specify detection thresholds application in specific diseases and time based on the needs.  In consultation with MoHZ, Epidemic Early Detection (EED) thresholds were established and customized within eIDSR. The EED module with the system is expected to utilize the threshold to trigger alerts when certain disease cases are reached as reporting continues. The system currently covers reporting from HF levels, However, plans are underway to extend this capability through the implementation of the EBS system to also cover data reported from the community level. In collaboration with MoHZ, MDH, CDC, and other key stakeholders, the eIDSR system is currently deployed and operational in Zanzibar since January 2023 covering around 370 Health facilities with automated early disease detection. The generification approach developed that strongly emphasizes configurations, has in a way proven to handle complexity in contextualizing settings for early detection operations. This approach has potential and can be contextualized in settings of other countries implementing disease surveillance systems.