A patient collapses in an emergency department three states from home, unconscious, with no family present and no way to describe a medication history that includes a blood thinner and a drug allergy that could kill her if the wrong antibiotic is given. Fifteen years ago, that emergency physician had almost nothing to go on. Today, in a growing number of communities, that physician can query a regional or national health information exchange network and pull up a medication list and allergy record from a hospital hundreds of miles away, in the time it takes to order a lab test. That capability did not arrive as one technology. It arrived as several distinct exchange models, built for different clinical situations, layered on top of each other over more than a decade of federal investment, regional experimentation, and increasingly, private-sector network building.

Health information exchange, commonly abbreviated HIE, refers both to the electronic movement of health-related information according to nationally recognized standards and, confusingly, to the organizations that make that movement possible. Understanding it requires separating two things that get discussed together but are not the same: the technical models by which information moves (directed, query-based, and consumer-mediated exchange), and the organizational layer operating the plumbing (state and regional HIEs, and the national networks now stitching those regional efforts together).

The Three Core Models of Health Information Exchange

The Office of the National Coordinator for Health Information Technology (ONC), the federal agency responsible for coordinating national health IT policy, describes three primary forms of exchange. Each solves a different clinical problem, and most provider organizations use more than one rather than picking a single approach.

Directed exchange: sending information to a known recipient

Directed exchange is the electronic equivalent of a certified letter. It lets providers send and receive secure patient information — laboratory results, referral summaries, or discharge instructions — directly to another known, trusted recipient over an encrypted connection, according to ONC’s guidance on health information exchange. The sender knows exactly who the recipient is before transmission, much as an email is addressed to a specific inbox.

The dominant implementation of directed exchange is the Direct Project protocol, often called “Direct secure messaging.” It layers standard email-like transport (SMTP) with security certificates that authenticate both organizations and encrypt the message in transit, so protected health information can move between two Direct-enabled systems without traveling as ordinary unencrypted email. A primary care physician referring a patient to a specialist, or a hospital sending a discharge summary back to a patient’s regular doctor, is a textbook use case: the sender knows the destination, and the transaction is fundamentally a push.

Directed exchange has been comparatively easy to adopt because it does not require a query infrastructure or a master patient index across organizations — it primarily requires that both parties have Direct addresses and trust each other’s certificates. That simplicity is a major reason it achieved broader early adoption than the other two models.

Query-based exchange: finding information you didn’t know existed

Query-based exchange (sometimes called “pull” exchange) lets providers find and request information about a patient from other organizations, typically where the treating clinician does not already know where prior records are held, according to ONC. This is the model the emergency department scenario above depends on: the physician does not know which hospitals or clinics hold this patient’s history, so directed exchange — which requires knowing the recipient ahead of time — cannot help. Query-based exchange instead broadcasts a request, checks an index of participating organizations for a match on patient identity, and retrieves whatever records those organizations are willing and able to share.

Query-based exchange is technically more demanding because it depends on several capabilities working correctly: a reliable way to match patient identities across organizations that may use different record numbers and demographic formats (a persistent challenge discussed below), participation agreements defining what data can be requested and when, and, at national scale, a common technical framework so a query from one network can be understood on a different network. It is most associated with unplanned or emergency care, but it is also used for planned care — a specialist pulling a patient’s recent imaging before a first appointment rather than asking the patient to hand-carry a disc.

Consumer-mediated exchange: the patient as the connector

Consumer-mediated exchange lets patients themselves aggregate their own health information from multiple providers, review it, correct inaccuracies, and control who else gets to see it, according to ONC’s framework. Rather than two provider organizations negotiating a data-sharing arrangement, the patient becomes the active party requesting records from each provider and, in more advanced implementations, authorizing an app or portal to pull that data on their behalf.

Patient portals tied to a single health system are a limited version of this idea, but true consumer-mediated exchange aggregates across organizations — a patient who has seen a primary care doctor, a cardiologist, and a hospital, each on a different EHR platform, viewing all three records in one place and deciding whether to share that combined view. It has generally lagged the other two models, both because it depends on the patient taking an active role and because it depends on providers exposing data through consumer-facing interfaces, historically a lower priority than clinician-to-clinician exchange. That is beginning to shift as newer data-sharing standards make it easier for a patient’s chosen app to pull structured data directly from a provider’s system, though broad adoption is still developing.

Regional and State HIEs: The Organizational Backbone

The three models describe how data moves. Regional and state health information exchange organizations historically built and operated the infrastructure making that movement possible within a defined geography.

Many trace back to Regional Health Information Organizations, or RHIOs — nonprofit, multi-stakeholder entities that emerged in the early-to-mid 2000s, often anchored around a metropolitan area or a state, bringing together competing hospital systems, independent physician practices, and sometimes payers to build shared infrastructure no single participant would have built alone. Federal funding under the HITECH Act of 2009, administered through ONC’s State Health Information Exchange Cooperative Agreement Program, accelerated this considerably, providing grants to all fifty states and several territories to establish or expand HIE capability.

That seed funding created a structural challenge that has shaped the landscape since: many state and regional HIEs built substantial infrastructure during the grant-funded years but faced a sustainability question once funding wound down. Some transitioned to fee-based models charging hospitals, practices, or payers; others consolidated with neighboring exchanges, merged into state-designated single entities, or ceased operation. Research tracking RHIO viability has found that organizations able to secure paying participants before becoming fully operational — a sign stakeholders saw clear value going in — were considerably more likely to remain financially sound than those relying primarily on continued grant support.

Today’s regional and state HIEs vary widely in structure. Some states designated a single statewide HIE as the authoritative exchange; New York’s Statewide Health Information Network, built from a federation of regional RHIOs, is a widely cited example. Others have multiple competing or geographically distinct HIEs within their borders. Regardless of structure, the core function is similar: onboarding local hospitals, practices, labs, and sometimes public health departments and payers, and providing both directed and query-based exchange among them.

National Networks: Connecting the Regional Pieces Together

A patient who moves, travels, or is treated outside their home region exposes the central limitation of a purely regional model: a query sent to a Boston-area HIE does nothing to surface a record in a Phoenix hospital’s system. Closing that gap has driven a set of national-scale initiatives that, as of 2020, are reshaping how exchange happens above the regional layer.

eHealth Exchange

eHealth Exchange is among the oldest and largest health information networks in the United States, with roots in a federal initiative that predates most current national interoperability efforts. It functions as a network in the traditional sense: participating organizations — including federal agencies such as the Department of Veterans Affairs and the Social Security Administration, along with hospital systems, regional HIEs, and other health IT organizations — sign a common trust agreement and connect using shared technical specifications, allowing any participant to query or send information to any other under agreed-upon rules. Governance and operational support have historically come from the nonprofit Sequoia Project.

Carequality

Carequality takes a different structural approach. Rather than functioning as a single network that organizations join directly, Carequality describes itself as a framework — a common set of legal terms, technical specifications, and governance rules that allows other, independent health data networks to interoperate, according to the Sequoia Project, which also stewards Carequality. Under this model, an EHR vendor’s own network of connected customers, or a health information exchange, can adopt the framework and immediately exchange with every other Carequality-connected network, without negotiating bilateral agreements with each one. This “network of networks” design is intended to solve connectivity at scale.

CommonWell Health Alliance

CommonWell Health Alliance began as a vendor-led effort, with several major EHR companies forming a not-for-profit trade association to build shared services — including patient identity matching and record location — that member vendors could build into their products so data could move between different vendors’ customer bases. In 2018, CommonWell and Carequality implemented a formal connection, allowing organizations on CommonWell-enabled systems to exchange with the broader universe of Carequality-connected networks and vice versa, substantially expanding the reach of both.

Where this is heading

As of 2020, ONC has been developing the Trusted Exchange Framework and Common Agreement (TEFCA), a policy initiative directed by the 21st Century Cures Act intended to establish a single “on-ramp” governance structure that participants in eHealth Exchange, Carequality, CommonWell, and other qualified networks could adopt to interoperate under one common set of rules nationwide. TEFCA has not yet been finalized or implemented; ONC has published draft versions for public comment, and its eventual scope, governance, and timeline remain to be determined. Organizations planning exchange strategy should treat TEFCA as a significant forthcoming development to monitor, not a live requirement.

Benefits of Health Information Exchange

The case for HIE, across all three models, rests on a consistent set of benefits reflected in ONC’s materials and published health services research:

  • Reduced duplicate testing. When a clinician can see a lab panel or imaging study was already performed elsewhere, unnecessary repeat testing — and its cost and delay — can often be avoided.
  • Improved care coordination. Directed exchange supports smoother handoffs between primary care, specialists, hospitals, and post-acute facilities.
  • Better-informed emergency care. Query-based exchange gives emergency clinicians medication lists, allergies, and problem lists for patients who cannot self-report.
  • Fewer adverse drug events. Visibility into a fuller medication history reduces the risk of dangerous interactions or duplicate prescribing across providers unaware of each other’s orders.
  • Greater patient engagement. Consumer-mediated exchange, where implemented, lets patients hold a more complete, portable copy of their own record.

Barriers to Broader Adoption

Despite more than a decade of investment, HIE adoption and use remain uneven across the country, and the barriers are well documented in health IT research and ONC’s program history:

Patient identity matching

Different organizations use different identifiers, formats, and data entry conventions for the same patient, and the United States has no unique national patient identifier. Matching “Robert Smith” at one hospital to “Bob Smith” at another, when both may share a birthdate with unrelated patients, remains an unresolved challenge that directly affects the reliability of query-based exchange.

Cost and technical complexity

Building and maintaining interfaces between EHR systems, HIEs, and national networks requires ongoing investment that smaller practices, safety-net providers, and long-term-care or behavioral health organizations — many ineligible for the EHR incentive payments that drove earlier adoption — have struggled to fund.

Sustainability of HIE organizations themselves

As noted above, many regional and state HIEs built infrastructure with time-limited federal grant funding and have had to find durable revenue models — participant fees, state appropriations, or value-added services — to remain operational, with mixed success.

Trust and data governance concerns

Competing health systems have not always been eager to make patient data available to competitors, and providers, patients, and legal counsel continue working through questions about liability, consent, and appropriate use once data crosses organizational lines. Behavioral health and substance use treatment records carry additional federal confidentiality protections that add complexity.

Workflow integration

Even where exchange is technically available, clinicians report that poorly integrated query results — an unwieldy stack of documents rather than organized clinical data — can add administrative burden rather than reduce it, limiting the real-world benefit of participation.

The Bottom Line

Health information exchange is not a single system a hospital either has or lacks. It is a layered set of capabilities: directed exchange for known handoffs, query-based exchange for unplanned care, and consumer-mediated exchange for patient-controlled access, operating on top of regional and state HIE infrastructure that is itself increasingly connected through national frameworks such as eHealth Exchange, Carequality, and CommonWell. None of these layers has fully solved the underlying challenges of patient matching, sustainable funding, and organizational trust, and the forthcoming TEFCA framework is intended to further consolidate governance across networks in the years ahead. Understanding which model applies to a given scenario — and which organizational layer carries the data — is the starting point for evaluating any HIE strategy or vendor claim.

This article is intended for general informational purposes about health IT policy and infrastructure and does not constitute medical, legal, or clinical advice.

Frequently Asked Questions

What is the difference between directed exchange and query-based exchange?

Directed exchange sends information to a specific, already-known recipient, similar to a secure email — used for referrals or discharge summaries. Query-based exchange lets a clinician search for and retrieve records from organizations they didn’t know held relevant data, which is why it’s the model typically used in emergency or unplanned-care situations.

Is health information exchange the same as a regional HIE organization?

Not exactly. “Health information exchange” describes the electronic movement of data using standardized methods, while an HIE organization is the nonprofit or state-designated entity that builds and operates the infrastructure — onboarding, agreements, and technical connections — enabling that movement within a region or state.

What is the difference between eHealth Exchange, Carequality, and CommonWell?

eHealth Exchange and CommonWell are networks that organizations join directly under a shared trust agreement. Carequality is a framework, not a network itself — it lets independently operated networks, including CommonWell’s, interoperate with each other under common technical and legal terms rather than negotiating separately.

Why hasn’t consumer-mediated exchange caught on as quickly as the other models?

It depends on patients actively requesting, aggregating, and managing their own records across providers, and on providers exposing data through consumer-facing tools rather than only clinician-to-clinician channels. Both pieces have historically lagged behind provider-to-provider exchange, though patient-facing data access is an active area of ongoing federal policy attention.

What is TEFCA, and is it active yet?

TEFCA, the Trusted Exchange Framework and Common Agreement, is a national interoperability governance structure being developed by ONC under the 21st Century Cures Act. As of 2020, it remains in draft and public-comment stages and has not been finalized or implemented, so no organization is currently required to participate in it.