Diagnostic Imaging Interoperability RFI Shines a Spotlight on the Lack of Patient-Centered Longitudinal Health Records

On January 30, 2026, the Assistant Secretary for Technology Policy/Office of the National Coordinator for Health Information Technology (ASTP/ONC) issued a Request for Information (RFI) seeking public input on the potential adoption of diagnostic imaging interoperability standards and certification criteria under the ONC Health IT Certification Program.

This follows ASTP/ONC’s publication of the HTI-5 Proposed Rule on Health Data, Technology, and Interoperability in late December, as well as the January 29 release of the Draft U.S. Core Data for Interoperability Version 7 (USCDI v7) for public comment through Standards Bulletin 2026-1.

What You Need to Know

• RFI: Written comments on the RFI are due March 16, 2026. These will inform any future rulemaking aimed at improving how imaging data is accessed, exchanged, and used by health care providers and patients.
• HTI-5: Recall that written comments on the HTI-5 Proposed Rule (see related EBG blog post here) are coming up fast, on February 27, 2026.
• USCDI v7: Written comments on USCDI v7 are due April 13, 2026. Depending on the comments submitted, ASTP/ONC plans to release the final version in July 2026.

These developments arrive at a pivotal moment and invite stakeholders to reflect on where interoperability efforts in U.S. health IT policy have succeeded and where they may be falling short. These developments are especially critical as the United States works to move beyond fragmented records toward truly comprehensive longitudinal health data to support individualized care, precision medicine, and patient-centered care.

A Historical Frame: What the ONC Was Intended to Achieve

Established by Executive Order 13335 in April 2004, ONC was charged with accelerating the adoption and nationwide implementation of interoperable health information technology so that health information could follow patients seamlessly and securely throughout their care.

Crucially, the executive order envisioned that electronic health records (EHRs) would replace fragmented paper records and support longitudinal patient records rather than isolated clinical snapshots. It anchored the idea that health IT infrastructure should reduce medical errors, improve quality of care, and empower patients and providers alike with shared, complete clinical information.

Over the ensuing two decades, federal laws—including the Health Information Technology for Economic and Clinical Health (HITECH) Act, the Patient Protection and Affordable Care Act (ACA), the Medicare Access and CHIP Reauthorization Act (MACRA), and the 21st Century Cures Act—have shaped the ONC’s Health IT Certification Program and helped catalyze widespread EHR adoption. But in practice, a significant byproduct of the certification regime was the rise of deep vendor-specific data silos, where systems can technically share data but often do so in siloed or proprietary ways that make seamless access to clinical data across systems elusive.[1]

The Imaging RFI: Opportunity or Risk of Re-Siloing?

The newly issued imaging interoperability RFI identifies persistent challenges in the exchange and accessibility of diagnostic imaging across the health care ecosystem, including reliance on portable media, limited Application Programming Interface (API) support, and inconsistent metadata tagging, among other issues. Diagnostic imaging data (the underlying pixels/data of CT scans, MRIs, and x-rays) is typically captured and stored in specialized systems outside the EHR, such as a Picture Archiving and Communication System (PACS) or Vendor Neutral Archive (VNA). Conversely, the textual radiology report created by the radiologist is stored directly within the EHR or in the Radiology Information System (RIS), which integrates with the EHR.

While the RFI reflects an attempt to address these challenges, there is a policy risk if imaging interoperability is pursued primarily through additional certification criteria. Absent careful alignment with broader interoperability frameworks, such an approach could unintentionally deepen data silos rather than integrate imaging into longitudinal health records. This risk is especially applicable to specialties such as oncology, cardiology, orthopedics, and emergency medicine. The RFI asks for an alternative, such as the adoption of standards and implementation mechanisms that integrate imaging into a unified clinical data layer, rather than erecting domain-specific boundaries that fragment information further.

HTI-5: A Broader Shift in Certification Strategy

As we’ve written previously, the Health Data, Technology, and Interoperability: ASTP/ONC Deregulatory Actions to Unleash Prosperity (HTI-5) Proposed Rule contemplates a fundamental recalibration of the ONC certification regime by:

• streamlining certification by removing redundant or stagnant criteria to reduce the burden on health IT developers;
• revising information blocking definitions and exceptions to support broader access, exchange, and use of electronic health information; and
• advancing a foundation for APIs based on Fast Healthcare Interoperability Resources (FHIR) that support modern interoperability and AI-enabled solutions rather than prescriptive, checklist-driven criteria.

HTI-5 represents a policy pivot away from compliance-heavy certification boxes toward standards that promote flexible and API-driven interoperability. That shift aligns more closely with the original intent of the ONC’s mission and recognizes that innovation often flourishes where standards enable, rather than mandate, behavior.

Standards, Models, and the Need for a True Clinical Data Layer

While certification has played an important role in standardizing baseline functionalities, the health data ecosystem has matured in ways that transcend old paradigms.

• FHIR offers standardized APIs that can support real-time, granular, and query-based exchange of clinical data, including imaging references.
• openEHR, which publishes technical standards for an EHR platform, is not new. On the contrary, it has been used for years internationally, e.g., by the National Health System in the UK and in Germany, Switzerland, Spain, and Australia.[2] What’s often missed in the United States is that openEHR and FHIR work best when used together.[3] openEHR provides the durable, longitudinal clinical data layer, while FHIR serves as the modern transport mechanism that moves data where it needs to go. Combined, they offer exactly what health care interoperability has been missing: persistence and portability.

These approaches, especially when paired, offer a pathway to interoperability that is more composable and clinically rich than traditional certification checklists alone. The fact that Health Level Seven International (HL7)—an American National Standards Institute-accredited standards organization dedicated to interoperability of health care data now has a CEO with an openEHR background underscores broader momentum toward these models as part of the interoperability evolution.

Thought leaders across the sector have similarly emphasized that consumer-centric, longitudinal records are foundational for data exchange and realizing the value of modern analytics and artificial intelligence in health care.[4] The RFI provides an opportunity for stakeholders to weigh in on the potential value of a clinical data layer for clinical research and innovation more broadly.

USCDI: Minimum Interoperability, Not a Longitudinal Record

As ASTP/ONC continues to evolve the USCDI, it is important to be clear about what the USCDI is and what it is not. USCDI establishes a policy-driven subset of data elements that certified health IT must be able to exchange and make accessible, largely to support the baseline required under the 21st Century Cures Act. Draft USCDI v7, in fact, includes 29 proposed data elements (e.g., diagnostic imaging reference, meaning the information that can be used to access a diagnostic imaging study, including an imaging study endpoint weblink, unique identifiers, and contextual information) and one significantly revised data element (tobacco use). Thirteen of the 30 proposed data elements are already largely supported for certified health IT. The full list is available in USCDI v7.

USCDI does not attempt to model the full breadth of clinical information generated during the course of care, nor is it designed to function as a comprehensive longitudinal patient health record. In practice, USCDI largely aligns with data treated as part of the designated record set under the Health Insurance Portability and Accountability Act (HIPAA) and HITECH, rather than encompassing all clinically relevant data generated across care settings and workflows. As a result, substantial portions of diagnostic, contextual, and workflow-derived clinical information remain outside its scope and locked in proprietary EHRs.

This limitation becomes especially apparent when compared to approaches such as openEHR. Whereas the USCDI focuses on what must be shared, openEHR is designed to model everything that is captured in a patient’s clinical journey in a computable, longitudinal form. An openEHR-based system can map selected elements to the USCDI and FHIR for regulatory exchange and patient access, but the reverse is not true: the USCDI alone cannot represent the richness, persistence, or clinical nuance required for a true longitudinal record.

As imaging interoperability policy evolves, this distinction matters. Expanding the USCDI or adding new certification criteria may improve access to specific data elements, but without a broader clinical data model, these efforts risk reinforcing a fragmented view of the patient record, one where compliance-defined subsets of data are interoperable, while the full clinical narrative remains locked in silos.

A Constructive Call to Action

ASTP/ONC’s imaging RFI offers a strategic opportunity for stakeholders to help shape a forward-looking policy framework that aligns imaging with HTI-5’s FHIR-forward future and accelerates progress toward a more interoperable, data-driven health care ecosystem. During this process, stakeholders can:

1. encourage ASTP/ONC to establish a voluntary pilot or demonstration program to evaluate end-to-end FHIR-based imaging exchange across unaffiliated systems—an initiative that could assess workflow integration, usability, and readiness for secondary uses such as research queries or de-identified analytics, while aligning with HTI-5’s deregulatory posture;
2. propose innovation-focused mechanisms such as challenge competitions coordinated with agencies—including the Centers for Medicare & Medicaid Services (CMS), the National Institutes of Health, or the Advanced Research Projects Agency for Health—to incentivize development of tools that enable hospitals, health systems, and value-based care entities to build clinical data layers above core EHR infrastructure without reverting to silos;
3. encourage ASTP/ONC to define success not only in terms of technical conformance, but also through measurable outcomes such as reduced redundant imaging, faster cross-organizational access to studies, and diminished reliance on custom point-to-point interfaces.

Such approaches could help align current federal policy efforts with the original ONC vision: rich, longitudinal, and patient-centered electronic health records that support continuity of care, clinical decision-making, and data-driven innovation.

Innovation and Research: Pulling all the Recent Regulatory Strings Together

Beyond clinical care, the absence of a longitudinal clinical data layer represents a significant missed opportunity for research and innovation. Clinical trials, real-world evidence generation, post-market surveillance, AI model development, and translational research all depend on high-quality, computable data across the patient journey. Yet much of this data remains fragmented across EHR modules, PACS, VNAs, registries, and proprietary vendor environments, forcing researchers and innovators to undertake costly extraction, normalization, and reconciliation efforts before meaningful analysis can even begin.

Through certification reform and refinement of information blocking policies (HTI-5), and innovation-oriented pilots (e.g., CMS and Medicaid Initiative) or challenge programs (e.g., the HHS EHIgnite challenge announced at the ASTP/ONC 2026 Annual Meeting), the ASTP/ONC has the opportunity to incentivize industry stakeholders to pivot towards what has long been understood to be a core objective: to advance longitudinal data persistence and computability to advance a patient-centered learning health care ecosystem.

Framed through the HTI-5’s deregulatory lens, advancing the adoption of a clinical data layer strategy should not be viewed as a push for the expansion of regulation. Rather, it is a structural simplification that lowers barriers to clinical and research innovation. Adopting durable, clinical data infrastructure can reduce duplicative integration costs, decrease vendor lock-in pressures, accelerate AI development, and shorten research timelines.

Nations that have invested in developing longitudinal, computable health data infrastructure have strengthened their positions in biomedical research and life sciences investment.[5] Without a comparable architectural evolution, the United States risks optimizing for compliance-defined minimum data exchanges while forfeiting leadership in data-driven medicine.

The radiologic imaging RFI should thus represent more than a domain-specific interoperability exercise. Radiologic imaging could serve as the foundation on which the modernization of the health IT certification program aligns with a broader strategy to introduce clinical data architecture that is flexible, vendor-neutral, contextually rich, and ready for research. This shift could bolster innovation and further support precision medicine, integrate laboratory and genetic data, advance safe, effective, and fair AI, introduce patient-generated data (such as data from remote monitoring devices), and further enhance other critical focus areas of U.S. innovators and biomedical researchers.

What Stakeholders Should Do Now

• Submit Comments on the RFI: Provide feedback on the imaging interoperability RFI by March 16, 2026, to influence future standards and ensure alignment with patient-centered, longitudinal health records.
• Assess Imaging Data Readiness: Evaluate your organization’s current imaging data exchange capabilities and identify gaps in interoperability, such as API support and integration with EHRs.
• Engage Technology and Compliance Teams: Collaborate with IT vendors and legal professionals to prepare for potential new standards and ensure compliance with evolving interoperability policies.
• Monitor Policy Developments: Stay updated on ASTP/ONC’s next steps, including potential pilot programs or new rulemaking, to align your organization’s strategy with future requirements.

* * * *

For additional information about the issues discussed in this Insight, please contact the attorney(s) listed on this page or the Epstein Becker Green Health Care and Life Sciences attorney who regularly handles your legal matters.

Epstein Becker Green Staff Attorney Ann W. Parks contributed to the preparation of this Insight.

ENDNOTES

[1] This pattern was anticipated as early as 2005, when some of the authors participated in a project sponsored by ONC, contract Number HHSP23320054100EC, and later published as Crime and Punishment: Can a NHIN Reduce the Cost of Health Care Fraud, predicting that national interoperability efforts would advance in uneven cycles rather than through a single, continuous implementation. Journal of Healthcare Information Management 22(3): 42–51. June 2008.

[2] Allen, A. “Why openEHR is Eating Healthcare.” Medium, 7 Jan. 2022, medium.com/@alastairallen/why-openehr-is-eating-healthcare-e28bd792c50c.

[3] Pedrera-Jiménez, M., et al. “Can OpenEHR, ISO 13606, and HL7 FHIR Work Together?” J Med Internet Res., 2023 Dec. 28;25:e48702. https://pmc.ncbi.nlm.nih.gov/articles/PMC10784985/.

[4] See examples of scholarly articles published in peer-reviewed journals that support the proposition that patient-centered care relies on valid, accurate, and complete data:

1. Consumer Data is Key to Artificial Intelligence Value: Welcome to the Health Care Future
2. Good data powers better health: Creating a longitudinal patient record with connected, meaningful data
3. Advancing Responsible Healthcare AI with Longitudinal EHR Datasets

[5] See Better’s Mission vol. 3, 2026, building health care on a foundation of trustworthy information, describing shared care records in Ireland, Greece, Slovenia, France, Spain, Indonesia, and others: https://3733374.fs1.hubspotusercontent-eu1.net/hubfs/3733374/2026/Mission_2026_Shared%20%20Care%20Record.pdf.