Standards and Interoperability - Condensed Notes

Learning Objectives

  • Define and describe the importance of standards and interoperability.
  • Discuss important identifiers in clinical information and challenges for patient identifiers.
  • Describe major transaction standards.
  • Discuss major messaging standards used in healthcare.
  • Describe key terminology standards in biomedicine and health.
  • Discuss interoperability requirements to emerge from the 21st Century Cures Act, including SMART on FHIR.

Core Concepts of Standards and Interoperability

  • Data standards promote consistent naming of patients, events, diagnoses, treatments, and more.
  • Four major types of healthcare standards:
    • Identifier standards (people/organizations, especially patients)
    • Transaction standards (regularize business interactions)
    • Messaging standards (move data between systems)
    • Terminology standards (normalize terms/concepts)
  • Interoperability: Interoperability=the ability of two or more systems or components to exchange information and to use that information that has been exchangedInteroperability = \text{the ability of two or more systems or components to exchange information and to use that information that has been exchanged}
  • 21st Century Cures Act provisions to increase interoperability among clinical data systems and EHRs (definition of interoperability).
  • The law emphasizes secure exchange, complete access/exchange/use of health information for authorized use, and prohibition of information blocking.
  • The Health IT landscape gained momentum with the HITECH Act, incentivizing widespread EHR adoption; standards adoption followed but varied.

Standards Development Overview

  • Hammond outlines four common approaches to standards development:
    • Ad hoc groups with informal specifications
    • De facto single vendor controls industry
    • Government mandate
    • Consensus among interested parties in an open process
  • Stages of standard development: Identification,Conceptualization,Discussion,Specification,Earlyimplementation,Conformance,CertificationIdentification, Conceptualization, Discussion, Specification, Early implementation, Conformance, Certification
  • Standards Development Organizations (SDOs):
    • ANSI (accredits SDOs)
    • ASC X12 (business transactions)
    • HL7 (messaging standards)
    • ASTM (health IT standards)
  • International bodies: ISO (health informatics through TC 215),  CEN TC 251 (health informatics)ISO\text{ (health informatics through TC 215)},\; CEN\text{ TC 251 (health informatics)}; ITU (telecom)
  • US governance and guidance: ONC (Interoperability Standards Advisory, ISA); NIST support; NLM terminology focus; IHE (non-federal interoperability showcases)
  • ONC ISA identifies interoperable standards for industry use.

Identifier Standards

Patient Identifiers

  • Purpose: unique linkage of records across organizations; reduces duplicate/overlaid records and supports data exchange.
  • Benefits vs. risks: easy linkage can threaten privacy/confidentiality.
  • Key attributes of identifiers (Riplinger): extUnique,Non-disclosing,Permanent,Ubiquitous,Canonical,Invariableext{Unique}, \text{Non-disclosing}, \text{Permanent}, \text{Ubiquitous}, \text{Canonical}, \text{Invariable}
  • National health identifiers are common outside the US (e.g., New Zealand NHI, Singapore NRIC). In the US, political barriers exist to a government-issued national health ID.
  • SSN as a de facto patient identifier has significant issues: duplicates, no check digit, cross-use for non-health purposes, de-identification risks.
  • Alternatives: probabilistic matching using attributes (name, address, DOB, phone, etc.); data quality and standardization are critical.
  • ONC has pushed for standardized patient identifiers and improved record matching; Project US@ standardizes addresses (uppercase, abbreviations, etc.).

Other Identifiers

  • Providers: National Provider Identifier (NPI) with a 10-digit number and a check digit.
  • Employers and health plans: Employer Identifier Number (EIN); Health Plan Identifier (HPI) or Other Entity Identifier (OEID) for transactions.
  • Facility identifiers help locate where care occurs.

Transaction Standards

  • ASC X12N standards support electronic health care business data exchange (e.g., claims, enrollments).
  • HIPAA mandated ASC X12N for administrative transactions; version 50105010 is the current baseline for many government and private payers.
  • Major transactions (examples):
    • Health claims and encounter information: 837837
    • Enrollment/disenrollment: 834834
    • Eligibility: 270/271270/271
    • Payment/remittance: 835835
    • Premium payments: 820820
    • Claim status: 276/277276/277
    • Referral/certification/authorization: 278278
    • Coordination of benefits: 837837
  • Notable ACA impact: 834 enrollment/disenrollment standard rollout highlighted implementation challenges during healthcare.gov launch.

Message Exchange Standards

  • HL7 is the umbrella organization; HL7 standards include Version 2 (V2) and Version 3, CDA, and more.
  • HL7 Version 2:
    • Widely used syntactic standard; ASCII-delimited messages; segments like MSH, PID, OBR, OBX.
    • Backward compatible across releases; primarily syntactic with limited semantics.
  • HL7 Version 3:
    • Semantic interoperability via Reference Information Model (RIM); XML-based; more complex; slower adoption.
  • HL7 CDA: Clinical Document Architecture; structured document templates; used for Continuity of Care Documents (CCD) and consolidated CDA (C-CDA).
  • Other messaging and data exchange standards:
    • DICOM for imaging data (headers + image data; PACS integration).
    • NCPDP SCRIPT for electronic prescribing.
    • CCR/CCD history as patient summaries; CCD based on CDA; Blue Button initiatives for patient access/export.
  • FHIR (see SMART on FHIR section) represents a modern approach bridging FHIR with other HL7 work.

Imaging Standards

  • DICOM standard defines transport of medical images and metadata; header contains patient, study, and device information; image data carried separately.
  • Challenges arise when non-DICOM transfers occur, risking incomplete or inconsistent image-related data in the patient record.

Prescribing Standards

  • SCRIPT (NCPDP) governs electronic prescribing communications between prescribers and pharmacies; mandated in Meaningful Use contexts.

Patient Summaries

  • CCR and CCD aimed to provide concise, portable patient summaries for care transitions.
  • CCD (C-CDA) represents a standardized, reusable approach to assembling document templates (e.g., allergies, medications, problems).
  • Blue Button initiatives enabled patient access to electronic summaries; Blue Button 2.0 expands to beneficiaries with CMS.

Terminology Standards

  • Purpose: semantic interoperability; normalize discrete data elements and support both structured data and narratives.
  • Key language concepts: vocabulary, concepts, terms, synonyms, polysemy, dictionaries, thesauri, vocabularies, ontologies.
  • Desiderata for vocabularies (Cimino): hierarchical structures, codes for concepts, and avoiding meaning-embedded codes; prefer purely identifier-based concept representation.
  • Major terminology families:
    • ICD-9/ICD-9-CM; ICD-10/ICD-10-CM; ICD-10-PCS (procedures)
    • DRGs (diagnosis-related groups) used for hospital reimbursement; aggregation of ICD codes into resource-based groups.
    • Drug terminologies: NDC, UNII, NDF-RT, NCIt SPL, RxNorm/RxTerms; FedMed reference terminology set.
    • LOINC for observations; ROC/RRELMA for mapping local codes to LOINC; LOINC covers tests/measurements and panels.
    • CPT-4 for procedures; HCPCS levels; E&M coding for encounters.
    • SNOMED CT: comprehensive clinical terminology with >300k concepts, >1M descriptions, >1M relationships; multiaxial/compositional approach; licensing through IHTSDO.
    • UMLS: Metathesaurus (concepts linked across vocabularies), Semantic Network, Specialist Lexicon; facilitates cross-terminology mapping.
    • Nursing vocabularies and challenges with harmonization; ongoing reconciliation efforts.
  • Other vocabularies/activities: CDT (dental CPT-like), MeSH (literature indexing), UMD (devices), DSM (psychiatric terms), ICPC, UCUM, CDES for research data elements (PROMIS, RaDaR, PhenX).
  • UMLS limitations: one-to-one mappings in Metathesaurus, no unifying hierarchy, not extensible like SNOMED, terms come from source vocabularies only.
  • Harmonization emphasis: SNOMED CT, LOINC, RxNorm as core triad for semantic interoperability; conceptual view: LOINC asks the question (what is measured), SNOMED CT provides the answer (what is diagnosed/condition).
  • Commercial terminology services exist (IMO, Medcin, 3M HDD) to map, update and provision terminologies.

SMART on FHIR and 21st Century Cures Act

  • FHIR provides a modern, flexible standard for data exchange; HL7 is the standards body behind FHIR.
  • What FHIR is not: not a terminology standard; not a security standard; not a user interface standard.
  • FHIR core concepts:
    • Resources as the basic data building blocks (six types: clinical content, identification, workflow, financial, conformance, infrastructure).
    • Bundles group interacting resources; Profiles constrain resources for context; Extensions add data beyond base models; Conformance defines how profiles are implemented.
    • Bulk data support for exchange across many patients for population health, research, etc.
  • SMART on FHIR: Mandl & Kohane concept of an app platform on top of a data store using FHIR APIs; supports app development and interoperability.
  • Security and access: OAuth2 is widely used to authorize access to FHIR data; OpenID Connect for identity.
  • US Core Data for Interoperability (USCDI): mandated data set for interoperability and certification; includes core patient data elements such as notes, demographics, medications, vital signs, etc.; aims to be interoperable via SMART on FHIR APIs.
  • Cures Rule (March 2020) key elements:
    • Data should be usable/movable without special effort;
    • Access to all data elements;
    • Open APIs;
    • Prohibition of information blocking with defined exceptions for harm/privacy/security, licensing, and feasibility.
  • Export and portability: EHI exportable in computable formats to enable system-to-system data movement when patients or providers switch systems.
  • Practical impact: push toward vendor interoperability, patient access, and improved data sharing; ongoing work to close gaps in USCDI data coverage for research and public health.

Conclusion

  • Widespread EHR use underscores the need for data standards and interoperability.
  • The Cures Act drives progress via USCDI, open APIs, and SMART on FHIR, with attention to social determinants of health (SDOH) data.
  • Clinicians, informaticians, and vendors will continue implementing these standards to improve care delivery, data quality, research, and patient empowerment.

Quick Reference Points

  • Major current transaction standard: 50105010 with transactions like 837837, 834834, 270/271270/271, 835835, 820820, 276/277276/277, 278278, 837837 (COB).
  • Major messaging standards: HL7 V2 (syntactic, ASCII-delimited); HL7 V3 and CDA (semantic, XML); DICOM for imaging; NCPDP SCRIPT for prescribing; CCD/C-CDA for patient summaries.
  • Core terminologies to know: ICD-10-CM/PCS, CPT/HCPCS, SNOMED CT, LOINC, RxNorm; UMLS as a map between vocabularies.
  • USCDI and SMART on FHIR are central to current interoperability efforts under the Cures Act.