Interoperability in health care needs to be intelligent to be meaningful

For more than a decade, health care has been chasing interoperability. We’ve connected electronic health records, built application programming interfaces (APIs) and stood up national exchange frameworks designed to move patient data across systems. In many ways, that effort has worked and data move more freely today than they did 10 years ago.

But in 2026, it’s increasingly clear that connectivity alone hasn’t delivered what the industry hoped it would. Records may travel, but they still don’t reliably translate into insight. The problem health care faces now is whether the data they exchange can actually be used. And that gap between connection and usefulness is where interoperability begins to break down.

Over the past decade, the industry has poured energy into standards like Fast Healthcare Interoperability Resources (FHIR), a modern data exchange standard for health information created by Health Level Seven International to enable systems to share clinical and administrative data efficiently, and initiatives like the Trusted Exchange Framework and Common Agreement (TEFCA), a nationwide framework and governance structure designed to support secure, standardized health information exchange across disparate health information networks.

These efforts created a common technical foundation and significantly reduced the friction of moving information between organizations, but they were always focused on transport, not meaning (and meaning matters).

Even in 2026, clinicians routinely encounter records that arrive incomplete, inconsistently structured or stripped of context. Lab results are formatted differently across systems; diagnoses appear under competing codes; and critical details are buried in free-text notes. The result is that care teams spend valuable time reconciling information that was supposed to be interoperable in the first place.

The root of the problem is semantic inconsistency. Even when records are exchanged successfully, the same clinical concept can be represented in multiple ways depending on the source system. Context is lost, definitions drift, and what looks like interoperability on paper becomes manual interpretation in practice.

That friction shows up everywhere when clinicians double-check records before making decisions, when administrative teams reverify information for coverage and reimbursement, or when analytics teams spend cycles cleaning data instead of using them. In many cases, interoperability simply shifts work downstream.

These limitations are becoming harder to ignore as health care pushes deeper into artificial intelligence (AI) and automation.

What makes this moment different is that health care is no longer experimenting at the margins. Payers, providers and life sciences companies are actively embedding automation and decision support into core workflows. At the same time, regulatory pressure and economic reality are converging such that clinicians are being asked to do more with fewer resources, and tolerance for inefficiency is evaporating. In this environment, workflows break down when data arrive in a form that still requires human cleanup before action can be taken.

This is where traditional interoperability assumptions begin to fail. Exchange frameworks were designed to ensure access rather than guarantee that the information arriving at the point of decision is complete, consistent or trustworthy. In turn, organizations often find themselves technically interoperable but practically constrained, still dependent on human judgment to resolve ambiguity before acting.

AI makes this gap even more visible. Advanced analytics and automation depend on structured inputs, consistent definitions and data that can be trusted. When underlying data lack shared meaning or provenance, AI actually compounds the risks at hand. Models trained on inconsistent or poorly contextualized data can produce outputs that look precise but aren’t reliable, which helps explain why so many health care AI deployments struggle to move beyond pilots.

Some will argue that interoperability is already solving these issues and that standards, APIs and exchange frameworks simply need broader adoption. And to be fair, those frameworks are essential and are the reason health care has made as much progress as it has.

But they were never designed to solve for meaning.

The persistence of manual review, duplicated work and inconsistent outcomes suggests that interoperability (as it is currently implemented) is necessary but incomplete. What health care needs next is intelligent interoperability, which is a layer that normalizes clinical concepts, preserves context and links data back to verifiable sources so they can be trusted across workflows and usable by default.

Health care’s next leap forward will come from embedding intelligence into the data themselves to extend interoperability from a concept of access to one of access and meaning. Until the industry makes that shift, connected health care will remain exactly that: connected but not fully usable.

Mika Newton is a health care technology innovator and CEO of xCures, where he has led the company’s transformation into a SaaS platform delivering AI-powered precision medicine and real-time clinical decision support. With more than two decades of leadership experience spanning life sciences, health care analytics and evidence-based research, Mika is recognized for his ability to translate cutting-edge data solutions into practical tools that improve care delivery and accelerate research. He is a frequent speaker and panelist at leading health care and technology forums, where he shares insights on AI, data interoperability and the future of personalized medicine. Mika’s background includes senior executive roles at Doctor Evidence, Evidera and Archimedes, where he drove innovation and expanded market reach.