Why Your Interoperability Strategy Is Still Failing: The Answer Is Not FHIR Alone

The Hard Number That Should Scare You

96 percent. That's the share of U.S. non-federal acute care hospitals that had adopted ONC-certified health IT by 2021 (ONC / HHS, Report to Congress). And about 4 in 5 office-based physicians had a certified EHR. We've spent billions, and we've digitized everything. But here's the uncomfortable question: why does clinical data still flow like molasses?

The answer isn't that we lack technology. It's that we've been chasing the wrong standard. HL7 v2 has been around since 1987 and still runs more than 95% of U.S. healthcare organizations (HL7 International). FHIR, the modern RESTful API standard, is now the darling of developers. But if you think swapping v2 for FHIR will fix your interoperability problems, you're setting yourself up for a rude awakening.

What FHIR Actually Solves (and What It Doesn't)

FHIR is a huge step forward. It uses RESTful APIs and represents clinical concepts as discrete resources like Patient, Encounter, and Observation (HL7 International). It's designed for mobile, cloud-native apps. The US Core Implementation Guide, based on FHIR R4, defines the minimum constraints for patient data access in the U.S. (US Core Implementation Guide). That's all good.

But FHIR is a transport and structure standard, not a semantic one. It tells you how to move data, not what the data means. If one system sends a lab result as a LOINC code and another sends it as a free-text string, FHIR won't save you. The resource can carry either, but the meaning is lost without shared vocabularies.

That's why HL7 International itself says complementary terminology standards are essential: SNOMED CT for clinical terms, LOINC for lab identifiers, and ICD-10 for diagnostic classification (HL7 International). FHIR is the envelope, not the message.

The Semantic Layer: Where Interoperability Actually Lives

Let's get concrete. LOINC version 2.82, released in February 2026, contains 109,325 total concepts, including 66,861 laboratory concepts and 28,635 clinical concepts (LOINC). SNOMED CT is designated for use in U.S. federal systems, owned by SNOMED International, with NLM as the U.S. National Release Center (NLM). RxNorm provides normalized names for clinical drugs and can mediate between different pharmacy vocabularies (NLM).

These are the standards that make data computable. Without them, you're just shuffling PDFs. But here's the trap: many organizations adopt FHIR and assume they're done. They map a few fields to US Core profiles, call it a day, and then wonder why their population health analytics are garbage.

The truth is, interoperability is a layered problem. FHIR handles the plumbing. SNOMED CT, LOINC, RxNorm, and DICOM for imaging (DICOM Standard) handle the semantics. You need both. And that's not a tech problem—it's a governance and strategy problem.

Why Your Current Strategy Is Doomed (and What to Do Instead)

Here's the blunt advice: stop treating FHIR as the finish line. Start treating it as the foundation. You need a semantic layer that maps every concept to a standard vocabulary. If you're a hospital, that means adopting SNOMED CT for diagnoses, LOINC for labs, RxNorm for meds, and DICOM for imaging. If you're a vendor, build your FHIR resources to use those codes by default.

Look at the numbers: HL7 v2 is still used by more than 95% of U.S. healthcare organizations (HL7 International). You can't rip that out overnight. But you can start building a FHIR façade that exposes semantic data, even if your backend is legacy. The key is to map your v2 messages to FHIR resources with LOINC and SNOMED codes. That's the only way to make data useful across systems.

And don't forget the regulatory push. The 21st Century Cures Act made information blocking illegal, and the ONC's information blocking exceptions are now in effect (ONC / HHS, Information Blocking). TEFCA is building a nationwide network-of-networks (ONC / HHS, TEFCA). These are forcing the issue. But regulation can't force semantic alignment. That's on you.

Bottom Line: Pair FHIR with Real Vocabularies—Now

If you take one thing from this, it's this: FHIR alone won't deliver interoperability. You need to pair it with SNOMED CT, LOINC, and RxNorm. Map everything to those standards, and you'll have data that actually means something. Skip that step, and you're just moving 96 percent of the problem around in a prettier box.

The best move is to start a pilot today: pick one clinical domain—say, lab results—and build a FHIR endpoint that returns LOINC-coded observations. Measure the difference in data quality. Then expand. It's not glamorous, but it's the only way to make interoperability real.

Sources

• ONC / HHS (Report to Congress) - https://healthit.gov/news/onc-outlines-health-it-interoperability-progress-report-congress/
• HL7 International - https://www.hl7.org/fhir/
• LOINC - https://loinc.org/
• NLM (SNOMED CT) - https://www.nlm.nih.gov/healthit/snomedct/index.html
• NLM (RxNorm) - https://www.nlm.nih.gov/research/umls/rxnorm/index.html
• ONC / HHS (TEFCA) - https://www.healthit.gov/topic/interoperability/policy/trusted-exchange-framework-and-common-agreement-tefca