Health Information Exchange (HIE)

A network model where clinical data remains stored at its source system within each participating organization. The HIE acts as a query broker rather than a data repository. When a clinician requests a patient's records, the HIE locates the data across the network and retrieves it on-demand.

• Data Custodianship: Each organization retains full control and legal responsibility for its own data
• Query Patterns: Uses IHE Cross-Community Access (XCA) profiles to locate and fetch documents
• Key Advantage: Reduces data duplication and medicolegal risk for data custodians
• Challenge: Requires all source systems to be online and responsive at query time

A model where clinical data from all participating organizations is aggregated into a single, unified clinical data repository (CDR). All queries are executed against this central store, which maintains a comprehensive longitudinal record for each patient.

• Data Flow: Uses ETL pipelines to extract, transform, and load data from source EHRs into the central repository
• Master Identity: Relies on an Enterprise Master Patient Index (EMPI) to link records across facilities
• Key Advantage: Fast, reliable queries independent of source system uptime
• Challenge: Significant data governance burden and potential for stale data if synchronization lags

A pragmatic model combining centralized and federated approaches. Frequently accessed data like medication histories and allergy lists are persisted in a central cache, while detailed clinical documents like operative notes are queried from source systems on demand.

• Caching Strategy: High-value, compact data types are normalized and stored centrally for rapid retrieval
• Document Retrieval: Bulky, infrequently accessed documents remain at the edge and are fetched via federated query
• Key Advantage: Balances query performance with data minimization principles
• Example: A statewide HIE caching lab results and immunizations centrally while querying radiology images from source PACS systems

The algorithmic backbone ensuring that clinical records from disparate systems are correctly linked to the right individual. Without robust identity resolution, an HIE cannot safely aggregate a longitudinal patient record.

• Deterministic Matching: Requires exact matches on a composite key of identifiers like Social Security Number and Date of Birth
• Probabilistic Matching: Uses statistical weighting on fields like name, address, and phone to calculate a match likelihood score
• Key Technology: An Enterprise Master Patient Index (EMPI) continuously cross-references and deduplicates identities
• Critical Metric: False-positive match rates must approach zero to prevent clinical data contamination

The policy engine and technical infrastructure that enforces a patient's legal right to control the disclosure of their Protected Health Information (PHI). Consent rules vary dramatically by state and data sensitivity.

• Granularity Models: Opt-in, opt-out, or granular consent by data type (e.g., block substance abuse records per 42 CFR Part 2)
• Enforcement Point: The HIE must intercept every query and filter results based on the patient's active consent directives
• Metadata Tagging: Clinical documents are tagged with sensitivity codes that the consent engine evaluates at query time
• Challenge: Maintaining accurate consent state when a patient revokes permission across a multi-node federated network

The immutable logging infrastructure that records every clinical data transaction across the HIE for security monitoring and compliance. These logs provide non-repudiation, proving definitively that a specific user accessed a specific record at a specific time.

• ATNA Profile: Uses the IHE Audit Trail and Node Authentication (ATNA) integration profile to standardize audit record format
• Log Contents: Captures user identity, patient identity, document accessed, timestamp, and purpose of use
• Forensic Capability: Enables detection of break-the-glass events where a clinician accesses records outside their normal scope
• Retention: Audit logs must be preserved for the duration specified by state law and organizational policy, often 6-10 years

The highest level of interoperability where systems not only exchange data but also interpret its clinical meaning without ambiguity. This requires shared, standardized medical terminologies.

• Depends on code systems like SNOMED CT and LOINC
• Ensures a diagnosis of 'DM' in one system is understood as 'Diabetes Mellitus' in another
• Contrasts with syntactic interoperability, which only guarantees structural compatibility

A centralized database that maintains a unique identifier for every patient across a healthcare organization, linking disparate medical records to prevent duplicate entries.

• Uses deterministic matching (exact field matches) or probabilistic matching (statistical likelihood)
• An Enterprise Master Patient Index (EMPI) extends this across an entire health system or HIE
• Critical for accurate patient identification when aggregating records from multiple sources

Middleware software that acts as a central translation broker, facilitating message routing, data transformation, and connectivity between disparate healthcare systems.

• Mirth Connect is a widely adopted open-source example
• Implements a hub-and-spoke model to replace brittle point-to-point interfaces
• Often uses a Canonical Data Model to reduce the number of required field-level mappings

An IHE integration profile that defines a federated query and retrieve mechanism for locating and fetching patient clinical documents across connected remote communities.

• Enables a health information exchange to query other HIEs without a centralized data repository
• Built on web services standards (SOAP-based)
• Returns documents in formats like Consolidated CDA (C-CDA)

A HIPAA-compliant, encrypted email protocol that enables authenticated healthcare providers to securely send health information directly to a known, trusted recipient over the internet.

• Uses X.509 digital certificates for identity verification and encryption
• Designed for point-to-point push of clinical information, such as a discharge summary to a primary care physician
• Contrasts with query-based exchange models like XCA

The administrative and technical system for capturing, storing, and enforcing a patient's granular permissions regarding the collection, use, and disclosure of their protected health information.

• Must align with federal (42 CFR Part 2) and state-specific privacy regulations
• Can implement opt-in, opt-out, or granular data element-level restrictions
• A foundational governance requirement for any multi-stakeholder HIE