This community innovation has been accepted at the 2025 DHIS2 Annual Conference
Enhancing Health Analytics with DHIS2 & FHIR
In the digital health landscape, the ability to share and analyze data seamlessly is crucial for improving health care, enhancing operational efficiency, and driving innovation. DHIS2 is one of the most used health information systems around the world that allows to track medical records, while FHIR (Fast Healthcare Interoperability Resources) is a standard developed by HL7 for storing and exchanging electronic health records, simplifying data sharing by using a consistent data model and API-based approach.
There are several scenarios where an integration between DHIS2 and FHIR are desirable, especially for automating data exchange for national health programs, data interoperability between different systems used by international health organizations, among others. Moreover, through the SMART Guidelines, WHO seeks to leverage FHIR as mechanism to standardize digital health interventions such as case management, immunization records, and disease surveillance.
While DHIS2 presents several powerful tools in order to analyze its data, analyzing FHIR based data might have different directions or approaches regarding: type of data handled, granularity of the analysis, security, real time vs, batch data processing, just to name a few. However, even if there are several organizations implementing FHIR data models, analyzing data based and stored on FHIR systems presents several challenges:
- Data complexity: FHIR resources can have complex, nested structures, making it difficult to parse and analyze the data effectively. Navigating these hierarchical JSON structures requires advanced querying techniques and tools.
- Performance issues: Large FHIR payloads can be cumbersome to process, leading to performance bottlenecks. Efficiently handling and querying large datasets requires optimized data processing strategies and robust infrastructure.
- Security and privacy: Ensuring compliance with regulations like HIPAA and GDPR while analyzing FHIR payloads is crucial to protect patient privacy.
- Data transformation: Converting JSON FHIR payloads into other formats for analysis or integration with different systems requires robust data transformation processes.
Google Data FHIR Pipes are a set of tools provided by Google to facilitate the integration, processing, and analysis of healthcare data based on FHIR standards. They are open source, based in Spark, can be executed in local setups, and they provide several features for data ingestion, transformation, scalability, advanced analytics and machine learning. They provide robust security measures and compliance with standards like HIPAA and GDPR to ensure patient and health care data is protected.
SolidLines works with global NGOs that collect health information at the point of care or community levels, and use these data to render analyses and reports at local and global levels. We have used DHIS2, FHIR databases and Google Data FHIR Pipes in multiple project settings to track patients and clinical encounters. Moreover, we have been collaborating with Google in the design of a global architecture that works for our use cases while we also try to address the longstanding challenge of data silos and fragmentation in healthcare, especially relevant for global and complex organizations.
We would like to present the design of a proposed open source architecture that integrates DHIS2, FHIR, Google FHIR Pipes, and open source BI tools like superset. Based on our experience we would also like to share best practices and discuss different alternatives about analyzing data that is DHIS2, FHIR based or both, highlighting differences and pros and cons.
Primary Author: Carlos Tejo Alonso
Keywords:
DHIS2, FHIR, Integrated Analysis