This community innovation has been accepted at the 2025 DHIS2 Annual Conference
The Automated Surveillance Alert threshold program
DHIS2 is well suited for routine disease surveillance and early warning systems due to its flexible, customizable data models. Since 2020, Zambia has used the electronic Integrated Disease Surveillance and Response (eIDSR) system—a DHIS2 based platform—to collect aggregated data on suspected and confirmed cases of outbreak prone diseases. However, eIDSR has not been fully utilized as an early warning system due to its limitations in computing complex thresholds and sending structured alerts to affected districts. To address this, we developed an external application to enhance eIDSR with automated threshold calculations, streamlined alert management, and direct notifications. Using a human centered design approach, we worked closely with national and subnational teams to review historical eIDSR data and define alert thresholds for ten conditions. We then developed an external application that pulls weekly data, calculates thresholds using Python, and directly notifies affected districts when thresholds are surpassed to update a DHIS2 tracker program. Verified alerts are automatically entered into emergency management system (EMS). Standard operating procedures were developed to streamline workflows. The program is currently being piloted in Central and Southern provinces. A total of 29 surveillance officers have been trained, and all 27 districts in the pilot provinces have been enrolled. As of January 31, 2024, the program has generated 76 alerts, with 54 verified (71%). Of these, 18 were discarded, while 36 were confirmed as events. These rarly results demonstrate how enhancing existing digital platforms can potentially support timely detection and managemnt of public health events. It is anticipated that lesson learned from the pilot will inform strategies to improve coordination of surveillance and response activities at national and subnational levels.
Primary Author: Steven Nonde
Keywords:
DHIS2, eIDSR, disease surveillance, early warning system, outbreak detection, threshold calculations, automated alerts, public health response, digital health, Zambi