A patient presents at an emergency department. They’re a regular at a different hospital across town. Their complete medical history—medications, allergies, recent test results, surgical history—exists in a database 15 kilometres away.

Can the ED doctor access it? Sometimes. Maybe. It depends on which hospitals, which systems, whether someone thought to set up a data-sharing agreement, and whether the patient remembered to mention where they usually receive care.

This is the interoperability problem in Australian healthcare, and it’s 2026 and we still haven’t properly solved it.

What Interoperability Actually Means

When health IT people talk about interoperability, they’re talking about different systems’ ability to exchange and use patient information. Not just send data, but send it in a format the receiving system can actually interpret and use clinically.

There are levels to this:

Foundational interoperability means systems can exchange data at all. Hospital A can technically send something to Hospital B. This is the easy part, and most systems can do it.

Structural interoperability means the data arrives in a standardised format. It’s not just a PDF of the discharge summary—it’s structured data fields that another system can parse. We’re getting better at this, but it’s inconsistent.

Semantic interoperability is the hard one. It means the receiving system actually understands what the data means. If Hospital A codes a diagnosis one way and Hospital B uses different terminology, does the system recognise they’re referring to the same condition? Often, no.

Why This Is Still Hard

You’d think by 2026 we’d have figured this out. Every other industry has. Your bank can talk to other banks. Retailers can process payments from any card. But healthcare? We’re still working on it.

Here’s why it’s harder than it looks:

Legacy systems everywhere. Hospitals are running electronic medical record systems that were implemented 10, 15, even 20 years ago. Upgrading is expensive and risky. So they persist, often unable to communicate effectively with modern systems.

No single standard that everyone actually uses. Yes, we have HL7 FHIR (Fast Healthcare Interoperability Resources), which is supposed to be the solution. And it’s genuinely good. But adoption is partial, implementation varies, and plenty of systems predate it entirely. HL7 Australia is pushing for broader FHIR adoption, but it’s a slow process.

Commercial incentives aren’t aligned. EMR vendors don’t necessarily benefit from easy data exchange. If switching costs are high because getting your data out is difficult, that’s vendor lock-in. Some have gotten better about this under regulatory pressure, but the fundamental incentive problem remains.

Privacy and security concerns. Healthcare data is sensitive. Setting up data sharing means navigating privacy law, patient consent, security protocols, and organisational liability concerns. Every new connection is a compliance exercise.

Organisational complexity. Even when the technology could work, getting the governance, legal agreements, and operational processes in place across multiple health services is a bureaucratic nightmare.

What’s Actually Working

It’s not all doom and gloom. Some things are improving.

My Health Record has become more useful as adoption has increased. It’s not comprehensive—lots of providers still don’t upload to it, and lots of patients don’t know it exists—but it’s creating a basic layer of data availability that didn’t exist a decade ago.

Regional health information exchanges are gaining traction. In some areas, local hospitals and clinics have set up shared systems that let them access each other’s records. The Victorian Government’s health data sharing initiatives have made genuine progress on this.

FHIR adoption is accelerating. New systems are being built FHIR-native, and older systems are adding FHIR APIs. This is creating a common language for data exchange that’s far better than the patchwork of proprietary interfaces we used to deal with.

Point-to-point integrations between major health services are increasing. It’s not elegant—each connection is custom-built—but tertiary hospitals that refer patients to each other frequently are setting up direct data feeds that work reliably.

The Clinical Impact

Let’s talk about what this actually means for patient care, because that’s what matters.

Medication reconciliation is a huge issue. If a patient takes five medications and nobody at the new hospital knows what they are, there’s real risk of dangerous interactions or gaps in treatment. Interoperability solves this. When it works.

Duplicate testing wastes money and delays care. If the patient had a CT scan yesterday at another facility, can you access those images? Or do you repeat the scan? Proper data sharing eliminates unnecessary repeat tests.

Longitudinal care for chronic conditions requires information continuity. If a patient sees a GP, a specialist, and gets admitted to hospital, everyone needs the full picture. Fragmented records mean fragmented care.

Emergency situations are where the stakes are highest. An unconscious patient can’t tell you about their drug allergies or recent procedures. If that information is locked in another system, it’s not available when it matters most.

The Technical Solutions Emerging

On the technology side, there’s actually some promising work happening.

FHIR-based aggregation platforms pull data from multiple sources into a unified view. They don’t solve the source system problem, but they create a better interface for clinicians to access distributed information.

National infrastructure initiatives like the ADHA’s (Australian Digital Health Agency) work on standardised APIs are creating technical frameworks for interoperability. Whether health services actually implement them is another question, but the standards exist.

Cloud-native EMRs are entering the market with interoperability baked in from day one, rather than bolted on later. As older systems get replaced, this should gradually improve the baseline.

Patient-mediated exchange is an interesting approach—giving patients control over their data and tools to share it across providers. It shifts the interoperability problem from hospital IT departments to consumer applications, which might actually be easier.

What Needs to Change

Technology isn’t really the blocker anymore. We have the standards and the tools. The barriers are organisational, financial, and regulatory.

Funding models need to support interoperability. Right now, hospitals pay to implement data sharing but don’t directly capture the benefits. The health system as a whole saves money and improves outcomes, but the individual hospital just sees costs. Until funding reflects the system-wide value, adoption will lag.

EMR vendor behaviour needs stronger regulation. If vendors make data exchange difficult, there should be consequences. Certification requirements that mandate proper FHIR implementation would help. Some jurisdictions are moving this direction.

Privacy law needs to be clearer about data sharing for care purposes. Clinicians are often uncertain about what they’re legally allowed to share and when. Clear, permissive rules for sharing data within the context of patient care would remove this friction.

Governance has to get simpler. The process of setting up data sharing agreements between health services is too slow and too complex. Standardised agreements and streamlined approval processes would help.

The Reality Check

Here’s the truth: we’re not going to have perfect interoperability any time soon. The health system is too complex, the legacy infrastructure is too entrenched, and the organisational challenges are too hard.

But we can get to “good enough.” We can get to a state where most patient data is accessible most of the time when it’s clinically needed. That’s achievable in the next 3-5 years if the focus and investment are there.

It requires treating interoperability as a core infrastructure investment, not an optional nice-to-have. It requires regulatory pressure on vendors and funding models that reward data sharing. And it requires health services to prioritise this even when it’s not directly tied to service delivery.

The technical solutions exist. What we need now is the organisational and political will to actually implement them. Because the cost of not solving this—in duplicated tests, medical errors, and fragmented care—is far higher than the cost of fixing it.