DHIS2
DHIS2 (District Health Information Software 2) is an open-source platform for collecting, validating, analysing and presenting health data. It is developed by the Health Information Systems Programme (HISP) at the University of Oslo and is used as the national routine health information system in a large number of countries.
It is a platform, not an application: what a DHIS2 instance does is decided by its metadata.
Two data models
Aggregate data
Counts for a period and an organisation unit — "BCG doses given, Tokha Health Post, Shrawan". This is the classic HMIS reporting model.
Building blocks: data elements, category combinations, data sets, periods, organisation units.
Tracker data
Individual-level records — a person enrolled in a program, moving through program stages over time. This is what supports continuity of care, follow-up and cohort analysis.
Building blocks: tracked entity, program, program stage, program rules, working lists.
The two models coexist: tracker data can be aggregated into the same analytics tables that aggregate reporting feeds.
Core concepts
| Concept | What it is |
|---|---|
| Organisation unit | A node in the hierarchy — country, province, district, facility |
| Data element | The thing being measured |
| Indicator | A calculated expression, typically numerator/denominator |
| Data set | A form: which data elements, collected at what frequency |
| Validation rule | A consistency check applied to entered data |
| Analytics tables | Pre-computed tables that make dashboards fast |
Analytics are generated by a scheduled job. Data entered after the last run does not appear in dashboards until the job runs again — a routine source of "the numbers are missing" reports.
The apps
- Data Entry / Aggregate Data Entry — form-based entry
- Tracker Capture / Capture — individual records
- Data Visualizer, Pivot Table, Maps, Dashboard — analysis
- Maintenance — metadata configuration
- Import/Export, Data Administration, Scheduler, Users — administration
- Android Capture — offline-first mobile data capture
Custom apps can be built against the Web API and installed into the instance.
API and interoperability
DHIS2 exposes a comprehensive REST API (/api/…) for metadata and data, which
is how most integrations are built. For standards-based exchange it also
supports FHIR-based integration through adapters and integration middleware —
see HL7 FHIR, integration engines
and OpenHIE, where DHIS2 typically occupies the HMIS and
analytics role.
What decides success
Implementations rarely fail on software. They fail on:
- Metadata design. Rushed data elements and category combinations are extremely expensive to change once data exists. See metadata design principles.
- Organisation unit hierarchy. It must match how the health system is actually administered.
- Data quality routines. Validation rules and follow-up, not exhortation.
- Training and support. Especially for tracker, where recording order matters.
- Server capacity. Analytics generation is resource-intensive at national scale.
Related
- eRecord documentation — user manual and implementation guide for a DHIS2-based deployment
- Amakomaya HMIS platform
- Analytics tools, GIS
References
- DHIS2 — https://dhis2.org/
- DHIS2 documentation — https://docs.dhis2.org/