A Cleaner Chain of Command for Health IT

Administrative reshuffles inside HHS usually land as management trivia. This one does not. The department’s March 31, 2026 decision to return enterprise technology, data, and AI leadership to the Office of the Chief Information Officer while restoring the Office of the National Coordinator for Health Information Technology to a narrower health IT policy role arrives at a point when governance failures are no longer abstract. They now show up as delayed authorizations, incomplete records, uneven cybersecurity, and growing concern about how AI enters care delivery.

The case for the change is not that a simpler org chart is inherently better. The case is that federal health technology has outgrown ambiguous ownership. Standards, certification, data operations, cloud architecture, cybersecurity, and AI oversight are now too intertwined to be split across overlapping titles without slowing execution. That makes the real question less about where boxes sit and more about whether HHS can finally connect policy, operations, and accountability in a way that reaches providers and patients.

Governance Has Become Clinical Infrastructure

For years, ONC has been asked to carry two burdens at once: set national rules for interoperability and also absorb broader technology leadership functions. That duality never fit the actual work. ONC’s strength is regulatory and market shaping. Its HTI-1 Final Rule advances interoperability, updates certification, adds information blocking refinements, adopts USCDI Version 3, and sets transparency expectations for predictive algorithms in certified health IT. ONC also continues to oversee the Trusted Exchange Framework and Common Agreement, which is meant to remove barriers to nationwide exchange across providers, public health agencies, patients, and payers. Those are powerful levers, but they are policy levers. They are not substitutes for an enterprise operating model.

Returning the chief technology, AI, and data roles to OCIO implicitly acknowledges that distinction. Enterprise architecture, identity, cybersecurity, cloud services, and data governance are not side functions attached to policy. They are the delivery mechanism that determines whether policy is felt as friction or relief. If that backbone is fragmented, even well written interoperability rules become another compliance burden layered onto already strained health systems.

Interoperability Has Moved Into a Harder Phase

The easiest era of interoperability is over. The industry no longer needs more proof that records can move. It needs proof that data can arrive in time, in usable form, and inside the workflow where decisions are made. ONC Health IT Research & Analysis now shows that hospital engagement in all four interoperability domains rose to 76 percent in 2025. Sending records reached 96 percent, receiving 93 percent, and finding information 94 percent. Yet integration, which is the point at which exchanged data becomes clinically useful inside the record, stood at 79 percent and was essentially flat compared with 2023. (ONC)

That plateau matters more than another press release about data liquidity. Information that can be located but not cleanly integrated still leaves clinicians reconciling outside data, revenue cycle teams chasing documentation, and patients repeating histories that already exist somewhere in the system. The gap between exchange and integration is where cost accumulates. It is also where many federal promises about affordability either become tangible or collapse under operational reality.

Affordability Is an Interoperability Test

The cleanest argument for the HHS restructuring is not philosophical. It is financial. CMS has already moved from broad interoperability rhetoric to specific administrative pain points through its Interoperability and Prior Authorization Final Rule. That rule requires affected payers to implement and maintain FHIR based APIs for patient access, provider access, payer to payer exchange, and prior authorization workflows, while also shortening decision timeframes and requiring more specific denial reasons. In other words, federal policy is no longer satisfied with data sharing as a virtue. It is demanding data sharing that reduces burden in a measurable way.

That policy direction aligns with public frustration. KFF reported in February 2026 that 69 percent of insured adults view prior authorization as at least a minor burden, and 34 percent identify it as the single biggest noncost burden in getting care. Among insured adults with chronic conditions, the pressure is even sharper. When patients experience delays or denials, the result is not just dissatisfaction. It is missed time, added administrative work, and, in a meaningful share of cases, reported harm to physical, mental, or financial well being.

That is why the HHS realignment should be judged against operational outcomes, not organizational neatness. If OCIO can help create common data services, better identity management, stronger security controls, and reusable API infrastructure, ONC’s policies have a far better chance of showing up as fewer manual touches, faster adjudication, cleaner documentation, and lower avoidable cost. If not, data liquidity will remain an elegant phrase for a problem still being worked around by phone calls, portals, and fax fallbacks.

AI Cannot Sit Outside the Operating Model

The same logic applies to AI. HHS Artificial Intelligence Strategy materials published in late 2025 focus on innovation, governance, and public trust, while the Office of Management and Budget has instructed agencies to accelerate adoption with safeguards proportionate to risk. That federal direction makes a standalone AI office less interesting than a functioning AI operating model. Responsible AI in health care depends on shared data governance, cybersecurity, auditability, workforce controls, procurement discipline, and clear accountability for deployment. Those are CIO level concerns as much as policy questions.

The clinical literature points in the same direction. A recent JAMA summit report argued that AI tools can affect care, operations, and payment in ways that are highly dependent on interface design, user training, and local setting, not only on model performance. The article also calls for stronger evaluation infrastructure and incentives that support interoperable data standards. That is a useful corrective to the current tendency to treat AI governance as a matter of principles alone. In practice, governance fails or succeeds in the handoff between standards and implementation.

Public trust is even more fragile than policy leaders sometimes acknowledge. A JAMA Network Open commentary published in 2025 noted low public trust in how health systems might use AI responsibly, building on nationally representative survey findings. That makes structural clarity inside HHS more than an internal management preference. If the federal government wants broader adoption of AI enabled health care, it needs visible lines of responsibility for how data is governed, how tools are monitored, and how harms are identified before scale turns them into systemic failures.

What This Move Should Deliver

A successful separation of duties between ONC and OCIO could make federal health technology more credible, not because the mission has narrowed, but because each office can now be measured against work it actually controls. ONC should be judged by the strength of standards, certification, and market rules. OCIO should be judged by whether enterprise data, security, and AI capabilities become more coherent across HHS and easier for operating divisions to use. That division of labor is overdue.

Still, the burden of proof sits with execution. The next evidence of success should not be another title change, another framework, or another aspirational speech about the future of digital health. It should be visible progress on integrated data exchange, prior authorization burden, trustworthy AI oversight, and the speed with which federal policy becomes usable infrastructure. Until then, the most important change at HHS is not that leadership has been aligned. It is that the excuses for misalignment have been reduced.