HEALTH INTEROPERABILITYREVIEW

Move data. Preserve meaning. Prove the exchange.

Newsroom

Health data exchange, standards, and infrastructure intelligence

Reporting on nationwide exchange, FHIR implementation, certification policy, identity, consent, terminology, public-health connectivity, and the systems responsible for moving usable health information.

Data Infrastructure

AWS HealthLake transformation needs source-field provenance

AWS describes HealthLake as a managed FHIR persistence layer and says a data-transformation agent can convert legacy clinical documents into queryable FHIR resources. A converted resource still needs source-document identity, field mapping, profile and version, terminology, uncertainty, validation, and receiving-workflow evidence before it can be trusted for use.

Moxe data exchange needs request-to-use lineage

Moxe presents secure clinical-data exchange connecting payers and providers and transforming data for payment and operational uses. Buyers still need to reconstruct the request purpose, permitted scope, source selection, transformation, provenance, delivery, acceptance, and downstream use for every exchange.

Azure API for FHIR retirement needs cutover proof

Microsoft says Azure API for FHIR retires September 30, 2026. The deadline calls for tested migration and rollback evidence—not assumed successor equivalence.

The 2026 ISA is an interoperability reference—not a universal implementation mandate

The Office of the Assistant Secretary for Technology Policy describes the Interoperability Standards Advisory as a public reference for identifying and assessing standards and implementation specifications. Inclusion helps planners compare maturity and use cases, but applicability, required version, conformance criteria, and production obligation still come from the controlling program, rule, contract, certification, or exchange agreement.

A Health Gorilla reconciled chart must preserve conflicting source records

Health Gorilla presents a pipeline that finds, matches, translates, de-duplicates, reconciles, traces, and delivers multi-source health data. A unified chart can reduce review burden, but a selected value must not erase the competing records, transformation, confidence, time, and clinical context needed to judge whether it is appropriate for a particular use.

CMS API version advancement needs endpoint-by-endpoint compatibility evidence

CMS lists standards and implementation-guide versions for Patient Access, Provider Access, Payer-to-Payer, Provider Directory, and Prior Authorization APIs and describes conditions for using updated versions. Version advancement can reduce technical stasis, but each payer still needs evidence that the chosen endpoint contract, data, authorization, clients, and transition preserve required access.

A NextGen Mirth channel retry needs duplicate-control and acknowledgment lineage

NextGen presents Mirth Connect as an integration engine for routing, transforming, mapping, and exchanging healthcare data across disparate systems. Retrying a failed channel can restore delivery, but it can also duplicate a message or repeat a downstream action unless source identity, attempts, acknowledgments, and receiver reconciliation stay connected.

An eHealth Exchange query needs permitted-purpose and responder-scope evidence

eHealth Exchange operates a nationwide health-information network and is a Designated QHIN under TEFCA, with official materials describing query, document exchange, public-health, federal, and other services. A successful query can prove that a request traveled and produced a response, but it does not by itself establish that the purpose, patient match, responders, data classes, and resulting use were appropriate and complete.

A Redox routing decision is not clinical document reconciliation

Redox presents healthcare data connectivity, normalization, and automated document triage and routing. Delivering a document to a likely destination can reduce queues, but it does not prove patient identity, encounter fit, document completeness, clinical interpretation, filing, acknowledgment, or reconciliation by the accountable care team.

Health Language mappings need source codes and version history

Wolters Kluwer presents Health Language for managing evolving terminologies, normalizing health data, and supporting interoperability and data quality. A standardized target can improve exchange and analytics, but the original code, terminology editions, mapping rule, ambiguity, and authorized use must remain visible.

A Bamboo Health event alert needs a separate care-action record

Bamboo Health describes real-time notifications when patients experience care events, alongside patient-history, discharge, and transition products. The alert can create timely awareness, but it does not show that the right person received it, assessed its meaning, acted, reached the patient, or completed a safe transition.

HealthShare record deduplication should keep clinical disagreement visible

InterSystems describes HealthShare Unified Care Record as an aggregated, normalized, and deduplicated patient record built from multiple sources. That architecture can reduce fragmentation while still preserving conflicting assertions, source context, and the authority required to reconcile them.

CommonWell pairs a master person index with record location—but retrieval is not identity proof

CommonWell's official site presents a nationwide exchange platform with a Master Person Index, Record Locator Service, Data Broker, and Trust Network. Those services can support discovery and retrieval across connected organizations, but a returned document does not by itself prove that every identity attribute is correct, the record belongs to the intended person, or the data is fit for a clinical decision.

Epic offers record exchange, patient-directed APIs, and partner access—but those routes are not interchangeable

Epic's interoperability page describes Care Everywhere record exchange, FHIR-based patient-directed APIs, Community Link access, government connections, health-plan data, referrals, Share Everywhere, and public interface specifications. Each route serves a different actor and purpose, so buyers need route-specific authority, payload, identity, workflow, and acceptance tests.

Firely Server provides FHIR infrastructure—but a running endpoint is not conformance

Firely presents server, API, validation, and conformance tooling for FHIR implementations. An endpoint can accept and return resources while still missing required profiles, terminology, search behavior, authorization, provenance, or production workflow semantics.

Verato identity resolution links records—it does not normalize clinical meaning

Verato presents identity resolution as a way to connect people, organizations, and networks across systems. That identity layer can help decide which records belong together without establishing that local codes, units, document context, provenance, or clinical meaning are equivalent.

Google Cloud Healthcare API keeps FHIR, HL7v2, and DICOM as separate stores

Google Cloud's official overview places FHIR, HL7v2, and DICOM data in modality-specific stores with distinct APIs. One managed service can centralize operations without making the formats, meanings, consent rules, or clinical workflows interchangeable.

FHIR R4 defines an interoperability base—not an implementation guide

HL7's permanent R4 publication defines resources, formats, APIs, terminology, and conformance infrastructure. A production exchange still needs the applicable profiles, implementation guide, versions, partner agreement, security, and tested behavior.

HTI-4 puts electronic prior authorization in certified health IT—not clinical decision authority

ASTP/ONC's HTI-4 final rule adds and updates health IT certification criteria for electronic prior authorization alongside electronic prescribing, real-time prescription benefit, and related APIs. Certification evidence addresses defined technology capabilities; it does not make the software the clinical or coverage decision-maker.

HTI-1 algorithm transparency targets certified health IT—not model approval

ONC's HTI-1 rule creates transparency requirements for AI and other predictive algorithms that are part of certified health IT. The disclosure baseline supports user assessment; it is not a federal approval of a model or a local decision to use it.

The Direct Standard secures transport—not disclosure authority

DirectTrust's Version 1.3 standard profiles secure, authenticated push messaging to known recipients. Delivery does not establish authorization, semantic completeness, reconciliation, or clinical use.

USCDI v6 expands device identity beyond implantables

The published data set broadens Unique Device Identifier from implantable devices to medical devices generally, creating a wider exchange target without proving capture, mapping, or clinical use.

TEFCA entry can run through a QHIN, Participant, or Subparticipant

The Recognized Coordinating Entity describes a network-of-networks hierarchy in which organizations may connect directly to a QHIN or use a Participant or Subparticipant path, with contracts and roles preserved at each layer.

US Core keeps Must Support separate from data availability

In US Core 9.0.0, Must Support sets implementation capability for responders and requestors; it does not promise that every marked element will be populated in every resource or clinically complete.

FHIR R5 is a release—not a production migration verdict

The current published specification gives architects a version to assess, but adoption still depends on implementation-guide dependencies, partner support, artifact maturity, and conformance evidence.

PDex 2.1 makes payer data exchange a provenance test

HL7's current PDex guide maps payer clinical, claims, encounter, and prior-authorization data into FHIR exchange. Trust depends on preserving source mappings, versions, provenance, and reconciliation.

FHIR Bulk Data 3.0 turns export into an asynchronous job

HL7's current published guide separates export kickoff, status polling, completion manifest, files, and errors. A working endpoint is only the start; buyers need the whole job and its authorization boundary.