4.4 Healthcare Data Standards, Terminologies & Code Sets

4.4 Healthcare Data Standards, Terminologies & Code Sets

Role of Standards in Clinical Informatics

For health information systems to communicate effectively, they must operate upon agreed-upon standards. In health informatics, standards are categorized into two primary domains:

1. Data Exchange and Messaging Standards (Syntax): Define the structural rules, message formats, delimiters, schemas, and transport protocols for packaging and transmitting data between software applications (e.g., HL7 v2, HL7 FHIR, DICOM).
2. Terminology Standards and Code Sets (Semantics): Standardize the specific clinical vocabularies, concepts, and numeric codes contained inside message fields, ensuring unambiguous semantic interpretation across system boundaries (e.g., SNOMED CT, LOINC, RxNorm, ICD-10).

Without standardized vocabularies, an EHR receiving a free-text lab result for 'Glucose' would be unable to determine whether the test represents a fasting blood glucose, a 2-hour postprandial glucose, or a spinal fluid glucose.

---

Messaging and Data Exchange Standards

HL7 Version 2 (HL7 v2)

Developed by Health Level Seven International, HL7 Version 2 (HL7 v2) is the legacy workhorse of inpatient healthcare interoperability. Introduced in the late 1980s, it remains deployed in nearly 95% of U.S. hospital systems for internal interface engine routing.

• Syntax: Relies on a lightweight, pipe-delimited text format (ER7 encoding). Characters such as pipes (|), caret symbols (^), and ampersands (&) act as field and component delimiters.
• Message Structure: Composed of ordered Segments (3-letter identifiers such as MSH for Message Header, PID for Patient Identification, PV1 for Patient Visit, OBR for Observation Request, and OBX for Observation Result).
• Trigger Event Framework: Operates on real-time clinical events. For example, an Admission, Discharge, Transfer event triggers an ADT^A01 message; a pharmacy order triggers an ORM^O01 message; a lab result triggers an ORU^R01 message.
• The Flexibility Paradox: HL7 v2 allowed optional fields and vendor-defined custom segments (Z-Segments). While this flexibility aided local implementation, it resulted in massive variation between hospital interfaces, preventing plug-and-play interoperability.

HL7 Version 3 (HL7 v3)

HL7 v3 was designed to eliminate the ambiguity of Version 2 by introducing a formal, model-driven methodology built upon the Reference Information Model (RIM) and XML encoding.

• Syntax: Complex XML schemas.
• Industry Outcome: HL7 v3 proved excessively complex, rigid, and difficult to implement. It suffered from a steep learning curve and achieved low adoption for transactional messaging, though its underlying model influenced the Clinical Document Architecture (C-CDA).

HL7 FHIR (Fast Healthcare Interoperability Resources)

HL7 FHIR (pronounced 'Fire') is the modern, web-native standard designed to revolutionize health data exchange by leveraging technologies that power the modern commercial internet.

• Web Architecture: FHIR utilizes RESTful web APIs, standard HTTP verbs (GET to read, POST to create, PUT to update, DELETE to remove), and lightweight JSON or XML payloads.
• Modular Resources: Instead of exchanging massive, monolithic clinical documents, FHIR breaks health data down into discrete, modular building blocks called Resources (e.g., Patient, Observation, Condition, MedicationRequest, DiagnosticReport).
• Granular Querying: Developers can execute precise API requests against an EHR database (e.g., GET /Patient/12345/Observation?code=2339-0 to retrieve only the blood glucose history for a specific patient) without downloading an entire medical chart.
• SMART on FHIR: An open-source app platform that allows third-party developer applications to run seamlessly inside any compliant EHR workspace using OAuth2 security.

---

Medical Imaging Standards: DICOM

Digital Imaging and Communications in Medicine (DICOM)

DICOM is the universal international standard for managing, storing, printing, and transmitting medical imaging information and associated metadata.

• Scope: Governs images generated by radiology modalities, including X-rays, Computed Tomography (CT) scans, Magnetic Resonance Imaging (MRI), Mammography, and Ultrasound.
• Dual Data Structure: A DICOM file encapsulates both the high-resolution pixel image dataset and an extensive header containing metadata (patient name, MRN, study date, modality, slice thickness, exposure settings, and equipment parameters).
• Interoperability Role: Ensures that an MRI scanner manufactured by Vendor A can transmit images across an Ethernet network to a Picture Archiving and Communication System (PACS) managed by Vendor B, which can then be rendered on a diagnostic workstation built by Vendor C.
• WADO (Web Access to DICOM Objects): Enables web-based EHR applications to retrieve DICOM images directly via HTTP RESTful requests.

---

Standardized Clinical Vocabularies & Terminologies

SNOMED CT (Systematized Nomenclature of Medicine -- Clinical Terms)

SNOMED CT is the most comprehensive, multilingual, and scientifically validated clinical healthcare terminology in the world.

• Primary Scope: Designed explicitly for use within EHRs to capture granular clinical information—including diagnoses, clinical symptoms, surgical procedures, body structures, etiology, and assessment findings—at the point of care.
• Polyhierarchical Structure: Concepts are organized in a rich, multi-axial network connected by formal semantic relationships (e.g., 'Acute Appendicitis' IS-A 'Appendicitis', which IS-A 'Inflammatory Disease of Appendiceal Structure').
• Components: Built upon Concepts (unique numerical identifiers representing clinical ideas), Descriptions (human-readable synonyms), and Relationships (linking concepts logically).

LOINC (Logical Observation Identifiers Names and Codes)

LOINC is the universal standard for identifying health measurements, laboratory test observations, and clinical measurements.

• Primary Scope: Standardizes laboratory orders and results (e.g., serum potassium, hemoglobin A1c) as well as clinical observations (vital signs, ECG measurements, standardized assessment scales).
• The Six-Part LOINC Axis: Every LOINC code is defined across six distinct analytical axes:
  1. Component (Analyte): The substance measured (e.g., Glucose).
  2. Property Measured: The attribute (e.g., Substance Concentration).
  3. Time Aspect: Timing of measurement (e.g., Point-in-time vs. 24-hour collection).
  4. System (Specimen): Specimen matrix (e.g., Serum, Urine, CSF).
  5. Scale Type: Measurement scale (e.g., Quantitative, Qualitative).
  6. Method Type: Analytical technique used (e.g., Enzymatic, Dipstick).

RxNorm

Maintained by the U.S. National Library of Medicine (NLM), RxNorm provides normalized names and unique identifiers for clinical drugs.

• Primary Scope: Standardizes medication representation across disparate pharmacy management software, e-prescribing networks, and drug interaction databases.
• Normalized Concepts: Assigns a unique Concept Unique Identifier (CUI) to every clinical drug formulation based on its active ingredient, strength, and dose form (e.g., Acetaminophen 325 MG Oral Tablet).
• Interoperability Role: Maps proprietary drug codes from commercial databases (e.g., First Databank, Multum, Micromedex) to a common reference standard, facilitating accurate electronic prescribing and automated drug interaction checking.

ICD-10-CM and ICD-10-PCS

Developed by the World Health Organization (WHO) and adapted for the U.S. healthcare system by NLM and CMS:

• ICD-10-CM (Clinical Modification): Used across all U.S. healthcare settings for coding diagnoses, medical conditions, and mortality statistics (3 to 7 alphanumeric characters).
• ICD-10-PCS (Procedure Coding System): Used exclusively in U.S. inpatient hospital settings for coding inpatient surgical and medical procedures (7 alphanumeric characters).
• Primary Use Case: Billing reimbursement, administrative claims processing, epidemiological tracking, and public health mortality reporting. Note: ICD-10 lacks the deep clinical granularity required for bedside progress note documentation, which is why EHRs use SNOMED CT internally and map to ICD-10 for billing.

CPT (Current Procedural Terminology) and HCPCS

Maintained by the American Medical Association (AMA):

• CPT (HCPCS Level I): Standardized 5-digit numeric code set used strictly for reporting medical, surgical, diagnostic, and physician evaluation and management (E/M) services performed in outpatient and physician office settings.
• HCPCS Level II: Maintained by CMS to code non-physician services, durable medical equipment (DME), prosthetics, orthotics, and injectable medications not covered by CPT.

---

Mapping and Translation Between Terminologies

Because healthcare systems use different code sets for different operational purposes, informatics professionals deploy Terminology Mapping.

For example, when a physician documents a patient's diagnosis of 'Type 2 Diabetes Mellitus with Kidney Complication' during an encounter, the EHR captures the entry using a granular SNOMED CT concept code (422034002). When the encounter closes, the EHR terminology service automatically cross-maps that SNOMED CT concept to the corresponding ICD-10-CM billing code (E11.22). This dual-coding architecture allows clinicians to document with clinical precision while administrative systems receive compliant billing codes.