The Interoperability Problem
A patient is transferred from a district hospital to a tertiary care centre for specialist management. The receiving team has no access to the sending hospital's clinical records. They restart the diagnostic workup from scratch — repeating blood tests, imaging, and clinical assessments that were completed 48 hours ago. The patient is exposed to additional radiation, additional cost, and additional delay. This scenario plays out thousands of times daily in healthcare systems around the world.
The same problem occurs at every transition of care: from hospital to primary care, from primary care to specialist, from inpatient to community, from one country to another. The information that would enable safe, efficient care exists somewhere in a digital system — but it cannot be accessed by the clinician who needs it, at the moment they need it.
This is the interoperability problem. It is not primarily a technical problem — the technical solutions are available and increasingly mature. It is a problem of economics, incentives, governance, and organisational will. Understanding its roots is the first step towards fixing it.
Levels of Interoperability
The Office of the National Coordinator for Health IT (ONC) and the Healthcare Information and Management Systems Society (HIMSS) use a widely accepted framework that describes interoperability at four levels:
Foundational interoperability (Level 1): One system can transmit data to another, and the receiving system can store it — even if it does not understand or process the content. This is the baseline: data moves from A to B without error.
Structural interoperability (Level 2): Data is exchanged using a defined format and structure. The receiving system knows what each field is and can parse the message correctly. HL7 v2 and FHIR both operate at this level.
Semantic interoperability (Level 3): Data not only has a defined structure but carries consistent meaning across systems. A code for "type 2 diabetes mellitus" in one system means exactly the same thing in another, because both systems use the same clinical terminology standard (SNOMED CT, ICD-10, LOINC). Without semantic interoperability, two systems can exchange data they cannot meaningfully compare or combine.
Organisational interoperability (Level 4): The governance, policy, legal, and social frameworks that enable data sharing across organisations are in place. This includes data sharing agreements, consent frameworks, privacy policies, and the organisational trust relationships that permit one organisation to access another's patient data.
Most healthcare IT conversations focus on levels 1–3. Level 4 — organisational interoperability — is often the hardest to achieve and the least discussed.
Why Interoperability Fails
Vendor Lock-In and Proprietary Formats
EHR and HIS vendors have historically had financial incentives to make their systems difficult to connect to. A hospital whose patient data is stored in a proprietary format that only the vendor can export is locked into that vendor for every future contract renewal. Information blocking — practices that restrict the flow of health information — has been sufficiently prevalent that the US Congress addressed it directly in the 21st Century Cures Act, which prohibits information blocking by EHR vendors and healthcare providers with significant penalties for violations.
Proprietary data formats remain a significant interoperability barrier in markets outside the US, where equivalent regulatory pressure on vendors may not yet exist.
Inconsistent Use of Standards
Even where standards exist and are nominally adopted, inconsistent implementation creates barriers. HL7 v2 is nominally universal — but every vendor implements it slightly differently, with different optionality, different code systems, and different handling of edge cases. Two systems both described as "HL7 v2 compliant" may still require significant custom mapping to exchange data reliably.
FHIR R4 is more prescriptive, but implementation guides — country-specific or use-case-specific extensions to the base standard — add variation. An Australian FHIR implementation and a US FHIR implementation both use FHIR R4 but may use different profiles, different terminology bindings, and different security frameworks.
Lack of Shared Patient Identifiers
Matching a patient's records across different systems requires a reliable common identifier. In countries with national patient identification systems — the NHS number in England, the Aadhaar number linked to Ayushman Bharat in India, or similar national health identifiers — this is technically achievable. In countries without a national patient identifier, or in environments where the identifier is not reliably captured, patient matching depends on probabilistic algorithms that are inevitably imperfect. Duplicates and misidentifications are common.
Incentive Misalignment
Healthcare organisations may not share the incentive to share data. A hospital system with a large captive patient population benefits from keeping patient data within its own systems — patients who cannot take their records elsewhere are less likely to seek care elsewhere. In fee-for-service payment models, sharing data that enables patients to receive care from competitors is economically counterproductive.
Value-based care models — where providers are rewarded for outcomes across a population rather than for the volume of services delivered — change these incentives, since care coordination across organisations is necessary for effective population health management. The transition towards value-based care is one of the strongest drivers of genuine interoperability investment.
Data Governance Complexity
Who owns patient health data? Who is permitted to share it? With whom? Under what conditions? These questions have different answers in different legal frameworks, and organisations that are uncertain about the answers default to not sharing. Appropriate data governance frameworks — clear policies on data ownership, consent, and sharing agreements — reduce this friction significantly.
HL7 FHIR as the Path Forward
HL7 FHIR R4 is the closest thing to a universal interoperability solution that currently exists. Its RESTful API architecture, web-standard data formats, and rich ecosystem of implementation guides make it more accessible and more consistent than any previous healthcare data standard.
Key FHIR-based interoperability mechanisms include:
SMART on FHIR: Enables authorised third-party applications to access patient data from FHIR-enabled EHRs using OAuth 2.0. This is the foundation for patient-facing mobile apps, care coordination tools, and clinical decision support applications that can plug into any FHIR-compliant EHR.
FHIR Bulk Data Access: Enables population-level data export from FHIR servers — essential for analytics, population health management, and health information exchange. The US CMS Blue Button 2.0 programme and the CARIN Alliance's consumer data exchange frameworks use bulk FHIR data access.
CDS Hooks: Enables external clinical decision support services to be invoked at defined points in EHR clinical workflows, receiving patient context and returning structured recommendations.
International Patient Summary (IPS): A FHIR-based specification for a minimal patient summary document intended to support cross-border care coordination. The IPS is being adopted by the EU, Canada, and other jurisdictions as the standard for cross-border patient record sharing.
National Health Information Exchanges
Health information exchanges (HIEs) — organisations that facilitate the sharing of health information across organisational boundaries within a defined jurisdiction — are a critical part of the interoperability infrastructure in many countries.
In the US, regional HIEs and the emerging Trusted Exchange Framework and Common Agreement (TEFCA) provide infrastructure for cross-organisation data sharing. In the UK, the NHS Spine and the National Record Locator provide central infrastructure for sharing key data elements across NHS organisations. Australia's My Health Record is a nationally managed repository for key health information. Similar national infrastructure exists or is being developed in Singapore, Saudi Arabia, the UAE, and many other countries.
These national platforms typically use FHIR as their primary API standard, creating a consistent integration target for organisations seeking to participate.
Practical Steps for Improving Interoperability
For an individual healthcare organisation, improving interoperability is a programme of work, not a single project. Practical steps include:
Assess the current state: Map your current integration landscape — which systems connect to which, using which standards, and with what level of reliability. Identify the clinical workflows most affected by interoperability gaps.
Adopt FHIR as a strategic requirement: New system procurements should require FHIR R4 support as a non-negotiable capability. Existing systems should be evaluated for FHIR roadmap maturity.
Implement a Master Patient Index (MPI): If you do not have a reliable patient identifier shared across all your systems, implementing an MPI with robust matching algorithms is a foundational step.
Deploy an integration platform: An enterprise integration platform — InterSystems HealthShare, Azure Health Data Services, Google Cloud Healthcare API, or equivalent — provides the infrastructure to connect systems and translate between standards.
Participate in national HIE infrastructure: Where national HIE platforms exist, connecting to them amplifies your interoperability capability beyond what you can achieve independently.
Address data governance: Develop clear policies on data sharing, patient consent, and data use agreements. These organisational interoperability foundations are prerequisites for technical interoperability to deliver value.
Invest in clinical terminology: Consistent use of SNOMED CT for diagnoses, LOINC for laboratory observations, and RxNorm or SNOMED for medications is the foundation of semantic interoperability. Terminology services and mapping tools support consistent terminology use across systems.
The organisations that achieve meaningful interoperability do so through sustained, senior-sponsored programmes that address technology, governance, and culture simultaneously.
FZ Consulting LLP provides interoperability strategy, integration architecture, and health information exchange advisory services to healthcare organisations globally. Contact our team to discuss how to advance your interoperability programme.