This community innovation has been accepted at the 2026 DHIS2 Annual Conference and will be in abstract track/lightning talk.
Automating DHIS2 Data Extraction in 34 countries
Routine health data stored in DHIS2 are increasingly used for monitoring, research, and policy dialogue. This reliance has intensified as delays or cancellation of household surveys -as a result of changing global health funding landscape - have increased dependence on routine health information systems. However, large-scale regional analysis of DHIS2 data remains constrained by fragmented extraction workflows, country-specific indicator definitions, inconsistent metadata, and high analyst effort in multi-country initiatives.
This abstract presents the CD2030 DHIS2 Extractor, a custom-developed application designed to automate and standardize DHIS2 data extraction across heterogeneous national implementations. Developed within a multi-country immunization and RMNCAH program spanning 34 countries, the extractor replaces manual scripts with a reproducible, metadata-driven pipeline that supports scalable data access and analysis. The CD2030 DHIS2 Extractor leverages DHIS2 Web APIs to dynamically map and extract indicators, data elements, organisation units, category combinations, and disaggregations using predefined templates and configurable metadata rules. This enables consistent extraction despite structural differences between DHIS2 instances while preserving national indicator definitions and reporting hierarchies and supports extraction of large datasets over extended time periods. Extracted datasets are integrated into open-source analytical workflows using R-based analytics, including Shiny applications and automated reporting using Quarto. This supports reproducible analyses, cross-country comparability of maternal and child health outcomes, and faster generation of research- and policy-ready outputs beyond standard DHIS2 dashboards. The CD2030 DHIS2 Extractor is deployed at scale and used by regional analysts and country teams. By mapping over 60 MNCH indicators within a standardized template, the tool has reduced analyst effort and enabled comparable multi-country analyses across 34 countries.
Primary Author: Peter Kaberia
Keywords:
API, DHIS2, Maternal health, DataExtraction, CD2030, APHRC
3 Likes