At TIBU Health, we’re now serving tens of thousands of patients and partnering with clinics, insurers, and corporate wellness programs. As our ecosystem grows, interoperability comes up constantly: how do we exchange data with partners without building a custom integration for every single one?

The standard answer is HL7 FHIR (Fast Healthcare Interoperability Resources). But most African health startups don’t need full FHIR compliance on day one, and pursuing it prematurely can waste months.

What is FHIR and why does it matter?

FHIR defines standardised “resources” like Patient, Observation, Medication, and Encounter with consistent APIs. If everyone uses FHIR, systems talk to each other without custom mappings.

Global adoption is real, and Africa is catching up through initiatives like Africa on FHIR. But regional adoption still faces real challenges:

• Knowledge Gap: FHIR expertise is hard to find on the continent.
• Infrastructure: Unreliable internet and legacy systems complicate implementation.
• Cost: Small teams often can’t resource a full implementation.
• Regulation: Many African countries don’t yet have strict mandates pushing adoption.

Where to start: Use cases that matter now

We’ve focused on high-value interoperability needs rather than chasing total compliance:

1. Lab integrations: FHIR’s DiagnosticReport and Observation resources standardise how results flow into the EHR. We implemented this for HIV viral load and TB screening and it saved weeks of bespoke development.
2. National reporting systems: Many countries use DHIS2. A FHIR bridge between your EHR and DHIS2 enables automated reporting instead of manual data entry.
3. Insurance claims: Exchanging patient eligibility and claims via FHIR reduces errors and speeds up reimbursement - something our insurance partners have been asking for.

What to ignore (for now)

• Full FHIR conformance: We don’t model every piece of data as a FHIR resource. It’s not necessary at our stage.
• Complex FHIR profiles: We use international base profiles and plan to adapt as national profiles emerge in Kenya.
• Advanced features: SMART-on-FHIR and CDS Hooks are powerful but not urgent when we’re still working on getting lab results flowing cleanly.

Practical implementation advice

We started with a FHIR facade - kept our internal data model as-is and built a thin API layer that translates to FHIR only when external systems request it.

Tools worth knowing:

• HAPI FHIR: A Java server for FHIR APIs.
• OpenMRS FHIR module: If you’re on OpenMRS.
• Matchbox: For FHIR validation.

Early FHIR adopters are going to have a real advantage as governments push for digital health integration. The pragmatic move is to implement it where it solves an actual problem today, and ignore it everywhere else.