Study guide
3.2 Clinical Terminologies & Ontologies: SNOMED CT, LOINC & RxNorm
AHIMA Certified Health Data Analyst CHDA
Track your progress as you read
Finished sections get a check mark.
1 Introduction & CHDA Exam Overview
0.1 About the AHIMA CHDA Exam 0.2 How to Use This Study Guide
2 Chapter 1: Healthcare Delivery Systems & Revenue Cycle
1.1 Healthcare Delivery Settings & Organizational Structures 1.2 Healthcare Revenue Cycle Management & Reimbursement Models 1.3 Regulatory Reporting & Accreditation Agencies
3 Chapter 2: Health Informatics, Datasets & AI
2.1 Electronic Health Records (EHR) & Health Informatics Foundations 2.2 Healthcare Datasets, Benchmarks & Population Health 2.3 Analytics Hierarchy & Artificial Intelligence in Healthcare
4 Chapter 3: Clinical Terminologies, Classifications & Standards
3.1 Clinical Classification Systems: ICD-10-CM/PCS & CPT/HCPCS 3.2 Clinical Terminologies & Ontologies: SNOMED CT, LOINC & RxNorm 3.3 Interoperability & Data Exchange Standards: HL7 v2, CDA & FHIR
5 Chapter 4: Business Needs Assessment & Scoping
4.1 Stakeholder Engagement & Analytics Needs Elicitation 4.2 Defining Metrics, KPIs & Healthcare Value 4.3 Analytics Project Management & Analysis Planning
6 Chapter 5: Data Sourcing, Lineage & Collection Methods
5.1 Healthcare Data Sourcing & Lineage 5.2 Healthcare Data Collection Methods & Architectures 5.3 Clinical & Financial Data Extraction
7 Chapter 6: Data Cleansing, Transformation & Validation
6.1 Healthcare Data Quality Assessment 6.2 Data Cleansing & Anomaly Remediation 6.3 Data Transformation, Mapping & Validation
8 Chapter 7: Exploratory Data Analysis & Descriptive Statistics
7.1 Healthcare Data Types & Scales of Measurement 7.2 Measures of Central Tendency & Dispersion 7.3 Exploratory Data Analysis in Clinical Populations
9 Chapter 8: Inferential Statistics & Hypothesis Testing
8.1 Sampling Techniques & Probability in Healthcare 8.2 Hypothesis Testing Fundamentals 8.3 Parametric vs. Non-Parametric Tests
10 Chapter 9: Advanced Statistical Modeling & Risk Adjustment
9.1 Correlation & Regression Analysis 9.2 Risk Adjustment & Healthcare Severity Modeling 9.3 Survival Analysis & Model Evaluation
11 Chapter 10: Healthcare Epidemiology & Study Methodologies
10.1 Epidemiological Measures of Disease Frequency & Association 10.2 Healthcare Study Designs & Methodologies 10.3 Confounding, Bias & Health Equity Analytics
12 Chapter 11: Healthcare Data Visualization & Dashboard Design
11.1 Principles of Effective Healthcare Data Visualization 11.2 Selecting Appropriate Healthcare Visualizations 11.3 Executive & Clinical Dashboard Architecture
13 Chapter 12: Healthcare Performance Measurement & CMS Quality Programs
12.1 CMS Quality Programs & Value-Based Purchasing 12.2 Clinical Quality Measures & Benchmarking 12.3 Financial Analytics & Case Mix Index Analysis
14 Chapter 13: Communicating Insights & Narrative Analytics
13.1 Translating Complex Analytics for Diverse Audiences 13.2 Contextualizing Limitations, Assumptions & Uncertainty 13.3 Narrative Reporting & Data-Driven Decision Support
15 Chapter 14: Healthcare Data Governance Frameworks & Stewardship
14.1 Enterprise Data Governance Frameworks 14.2 Data Stewardship & Master Patient Index (MPI) Management 14.3 Metadata Management & Business Glossaries
16 Chapter 15: Healthcare Privacy, Security & Data Ethics
15.1 HIPAA Privacy & Security Rules for Analysts 15.2 De-Identification Methodologies 15.3 Data Ethics, Audit Trails & Secondary Data Use
3.2 Clinical Terminologies & Ontologies: SNOMED CT, LOINC & RxNorm
Key Takeaways Explain This Topic AI
Classification systems (ICD-10) pigeonhole clinical concepts into mutually exclusive administrative bins for billing, whereas reference terminologies and ontologies (SNOMED CT, LOINC, RxNorm) utilize polyhierarchical, logic-based concept networks for direct point-of-care EHR documentation and semantic computing.
SNOMED CT organizes clinical knowledge across Concepts (SCTID), Descriptions (Fully Specified Names and Preferred Terms), and Relationships (IS_A hierarchies and concept model attributes like finding site and causative agent), supporting both pre-coordinated codes and post-coordinated compositional grammar.
LOINC standardizes laboratory observations, diagnostic measurements, clinical documents, and survey instruments (PHQ-9, MDS) through a rigorous 6-part naming structure: Component/Analyte, Property Measured, Time Aspect, System/Specimen, Scale Type, and Method Type.
RxNorm normalizes generic and branded medications produced by the NLM into structured concepts—including Ingredients (IN), Clinical Drug Forms (CDF), Semantic Clinical Drugs (SCD), and Semantic Branded Drugs (SBD)—resolving the severe data instability caused by manufacturer-driven NDC turnover and repackaging.
The Unified Medical Language System (UMLS) bridges over 200 biomedical terminologies via the Metathesaurus (CUIs), Semantic Network (127 semantic types and 54 relationships), and SPECIALIST Lexical Tools.
Last updated: August 2026
Clinical Terminologies & Ontologies: SNOMED CT, LOINC & RxNorm
Modern healthcare analytics requires capturing clinical observations, diagnostic findings, laboratory measurements, and medication regimens with granular precision at the point of care. While administrative classification systems like ICD-10-CM and CPT are indispensable for aggregation and prospective billing, they are deliberately constrained by mono-hierarchical structures, administrative grouping rules, and mutual exclusivity. In contrast, reference terminologies and formal clinical ontologies —specifically SNOMED CT , LOINC , and RxNorm —provide expressive, computable, and polyhierarchical semantic representations that power Electronic Health Record (EHR) documentation, Clinical Decision Support (CDS) engines, and automated quality computation.
1. Classification Systems vs. Reference Terminologies & Ontologies
Health data analysts must grasp the fundamental architectural differences between administrative classifications and clinical reference terminologies: +---------------------------------------------------------------------------------------------------+
- | CLASSIFICATIONS vs. REFERENCE TERMINOLOGIES |
- +-----------------------------------+---------------------------------------------------------------+
- | ATTRIBUTE | CLASSIFICATION SYSTEMS (ICD-10, CPT) |
- +-----------------------------------+---------------------------------------------------------------+
- | Primary Purpose | Statistical reporting, morbidity aggregation, billing |
- | Granularity | Broad groupings; "pigeonholes" clinical entities into bins |
- | Hierarchy Structure | Mono-hierarchical tree (single parent per category) |
- | Clinical Expressiveness | Low; nuances lost to fit mutually exclusive rules |
- | Logical Inference Support | Minimal; cannot perform formal description logic reasoning |
- +-----------------------------------+---------------------------------------------------------------+
- | ATTRIBUTE | REFERENCE TERMINOLOGIES / ONTOLOGIES (SNOMED CT, LOINC, RxNorm)|
- +-----------------------------------+---------------------------------------------------------------+
- | Primary Purpose | Direct EHR clinical documentation, CDS, semantic computing |
- | Granularity | High; captures exact clinical findings, sites, and forms |
- | Hierarchy Structure | Polyhierarchical DAG (multiple parents / IS_A relationships) |
- | Clinical Expressiveness | High; supports post-coordination and compositional grammar |
- | Logical Inference Support | Robust; allows automated subsumption testing and classification|
- +-----------------------------------+---------------------------------------------------------------+
The "Pigeonhole" Problem: In a classification system, every clinical condition must fit into exactly one code category to prevent double-counting in morbidity tables. For example, a patient with pneumococcal lobar pneumonia in ICD-10-CM is assigned code J13 , placing it strictly within the respiratory system hierarchy. However, from an ontological perspective, this disease is simultaneously an infectious bacterial disease and a respiratory disorder .
Polyhierarchy in Ontologies: Reference terminologies represent knowledge as a Directed Acyclic Graph (DAG), where a single clinical concept can inherit properties from multiple parent concepts ( IS_A relationships). This allows an analyst querying for all "bacterial infections" OR all "pulmonary diseases" to automatically retrieve the exact same clinical cohort without manual crosswalk maintenance.
2. SNOMED CT (Systematized Nomenclature of Medicine -- Clinical Terms)
Maintained internationally by SNOMED International (International Health Terminology Standards Development Organisation [IHTSDO]) and distributed within the United States by the National Library of Medicine (NLM), SNOMED CT is the world's most comprehensive, multilingual clinical healthcare terminology. +---------------------------------------------------------------------------------------------------+
- | SNOMED CT TRI-PARTITE CORE MODEL |
- +-----------------------------------+-----------------------------------+---------------------------+
- | CONCEPTS (SCTID) | DESCRIPTIONS | RELATIONSHIPS |
- | - Unique 6-18 digit integer | - Human-readable clinical terms | - IS_A Hierarchy (DAG) |
- | - Represents single clinical idea | - Fully Specified Name (FSN) | - Concept Model Attributes|
- | - Concept status: Active/Inactive | - Preferred Terms (PT) & Synonyms | - Description Logic Links |
- +-----------------------------------+-----------------------------------+---------------------------+
- The Core Components of SNOMED CT
Concepts (Concept IDs):
A Concept is a unique, language-independent clinical idea identified by a 6- to 18-digit integer ( SCTID ). Example: Concept 22298006 represents Myocardial infarction .
Concept identifiers are designed for semantic stability; when a meaning becomes obsolete or materially changes, the concept can be inactivated and relationships or replacements maintained by the terminology release. If a concept becomes obsolete or clinically invalid, its status is set to Inactive , and historical data references remain intact.
Descriptions (Description IDs):
Human-readable terms associated with a specific Concept ID. Each Description has its own unique Description ID (SCTID).
Fully Specified Name (FSN): The unique, unambiguous name that clarifies the precise clinical meaning of the concept by appending a Semantic Tag in parentheses. Example: Myocardial infarction (disorder) distinguishes the pathological process from Structure of myocardium (body structure) or Myocardial infarction protocol (procedure) .
Preferred Term (PT): The common term preferred by clinicians in a given language or dialect for display in user interfaces (e.g., Heart attack or Myocardial infarction ).
Synonyms (SYN): Alternative terms, acronyms, and lexical variants linked to the concept (e.g., Cardiac infarction , MI ).
Relationships (Relationship IDs):
Formal logical statements that link concepts together within the ontology.
Hierarchical ( IS_A ) Relationships: Establish parent-child inheritance. Example: Concept 22298006 ( Myocardial infarction ) IS_A 57054005 ( Acute coronary syndrome ), which in turn IS_A 53741008 ( Coronary artery disease ).
Attribute Relationships (Concept Model): Define the clinical characteristics of the concept using standardized semantic roles:
- Finding site = 74281007 ( Myocardium structure )
- Associated morphology = 55641003 ( Infarct )
- Causative agent = 416098002 ( Ischemia ) ┌───────────────────────────────────────────────┐
- │ Concept: Bacterial pneumonia │
- │ (SCTID: 53084003) │
- └───────┬───────────────────────────────┬───────┘
- │ IS_A │ IS_A
- ┌───────────────▼───────────────┐ ┌───────▼───────────────────────┐
- │ Infective pneumonia (disorder)│ │ Bacterial infectious disease │
- └───────────────┬───────────────┘ └───────────────────────────────┘
- │ Finding Site
- ┌───────────────▼───────────────┐
- │ Lung structure (anatomy) │
- └───────────────────────────────┘
- Pre-Coordination vs. Post-Coordination
Pre-Coordination: Using a single, existing SNOMED CT Concept ID to represent a complete clinical statement (e.g., 371087003 = Fracture of neck of right femur ).
Post-Coordination (Compositional Grammar): Combining multiple SNOMED CT concepts using standardized syntax to represent complex clinical scenarios that do not exist as a single pre-coordinated code. This allows clinicians and analysts to express granular clinical detail without waiting for standards committees to release new static codes.
Example Expression: 284196006 | Burn of skin | : 363698007 | Finding site | = 7824000 | Structure of thumb | , 246112005 | Severity | = 24484000 | Severe |
3. LOINC (Logical Observation Identifiers Names and Codes)
Developed and maintained by the Regenstrief Institute , LOINC provides universal identifiers for laboratory tests, clinical observations, diagnostic measurements, document types, and standardized assessment questionnaires. +---------------------------------------------------------------------------------------------------+
- | LOINC 6-PART FULLY SPECIFIED NAME |
- +-----------+--------------------+---------------+-----------------+------------+-------------------+
- | 1. COMP. | 2. PROPERTY | 3. TIME | 4. SYSTEM | 5. SCALE | 6. METHOD |
- | Analyte | Measured Kind | Timing/Aspect | Specimen/Context| Precision | Technique (Opt.) |
- +-----------+--------------------+---------------+-----------------+------------+-------------------+
- | Glucose | MCnc (Mass Conc.) | Pt (Point) | Ser/Plas | Qn (Quant) | [Blank/Generic] |
- | LOINC: 2345-7 (Glucose [Mass/volume] in Serum or Plasma) |
- +-----------+--------------------+---------------+-----------------+------------+-------------------+
- | Hb A1c | MFr (Mass Fraction)| Pt (Point) | Bld (Whole Bld) | Qn (Quant) | HPLC |
- | LOINC: 4548-4 (Hemoglobin A1c/Hemoglobin.total in Blood by HPLC) |
- +-----------+--------------------+---------------+-----------------+------------+-------------------+
- | SBP | Pres (Pressure) | Pt (Point) | Arterial system | Qn (Quant) | Auscultation |
- | LOINC: 8480-6 (Systolic blood pressure) |
- +-----------+--------------------+---------------+-----------------+------------+-------------------+
- The Six Axis Dimensions of LOINC
Every LOINC code consists of a 3- to 7-digit integer plus a mod-10 check digit (e.g., 2345-7 ), mapped to six fundamental semantic axes:
Component / Analyte (Part 1): The specific substance, biomarker, organism, or physiological parameter being measured (e.g., Glucose , Troponin I , Potassium , Body weight , PHQ-9 Total Score ).
Property Measured (Part 2): The physical or chemical characteristic of the analyte:
- MCnc = Mass Concentration ( m g / d L mg/dL m g / d L , μ g / L \mu g/L μg / L )
- SCnc = Substance / Molar Concentration ( m m o l / L mmol/L mm o l / L )
- MFr = Mass Fraction (percentage of total mass)
- PrThr = Presence or Threshold (qualitative detected/not detected)
- Pres = Pressure ( m m H g mmHg mm H g )
- Type / Kind = Nominal or categorical classification
Time Aspect (Part 3): The temporal window over which the observation was gathered:
- Pt = Point in time (spot collection, instantaneous measurement)
- 24H = 24-hour aggregate collection (e.g., 24-hour urine protein)
- 8H , 12H = Timed interval collections
System / Specimen (Part 4): The biological matrix or anatomical specimen containing the analyte:
- Ser/Plas = Serum or Plasma
- Bld = Whole Blood
- Urine = Urine
- CSF = Cerebrospinal Fluid
- Patient = Whole individual (used for vital signs and survey scores)
Scale Type (Part 5): The mathematical scale and data precision of the reported result:
- Qn = Quantitative (numeric value with units, e.g., 120 mg/dL )
- Ord = Ordinal (ranked categories, e.g., 1+ , 2+ , 3+ or Trace/Moderate/High )
- Nom = Nominal (unranked categories, e.g., Staphylococcus aureus )
- Nar = Narrative (free-text clinical impression)
- Doc = Document (complete clinical document architecture/summary)
Method Type (Part 6 - Optional): Specific laboratory assay method or measurement technique when clinically significant (e.g., EIA = Enzyme Immunoassay, HPLC = High-Performance Liquid Chromatography, Automated count , Estimated from Jaffe method ). Omitted if the clinical interpretation is identical across standard methods.
Laboratory vs. Clinical LOINC
Laboratory LOINC: Standardizes clinical chemistry, hematology, serology, microbiology, and molecular genetics results transmitted via HL7 ORU^R01 messages or FHIR Observation resources.
Clinical LOINC: Identifies non-laboratory clinical observations:
Vital Signs: Blood pressure ( 8480-6 ), heart rate ( 8867-4 ), body temperature ( 8310-5 ), pulse oximetry ( 59408-5 ).
Clinical Documents: Progress note ( 11506-3 ), Discharge summary ( 18842-5 ), Consultation note ( 11488-4 ).
Survey Instruments & Screening Assessments: Patient Health Questionnaire-9 (PHQ-9) depression score ( 44249-1 ), CMS Minimum Data Set (MDS) items, OASIS functional items, and Social Determinants of Health (SDOH) PRAPARE screening instruments.
4. RxNorm (Standardized Medication Terminology)
Produced and maintained by the National Library of Medicine (NLM) , RxNorm is the standard normalized clinical drug naming system in the United States. It provides normalized names and stable concept identifiers ( RxCUI ) for clinical drugs and links them to commercial drug knowledge bases (e.g., First Databank, Medi-Span, Micromedex, Gold Standard). +---------------------------------------------------------------------------------------------------+
- | RxNORM CONCEPT DERIVATION HIERARCHY |
- +---------------------------------------------------------------------------------------------------+
- | IN (Ingredient): Metformin |
- | PIN (Precise Ingredient): Metformin hydrochloride |
- | DF (Dose Form): Oral Tablet |
- | CDF (Clinical Drug Form): Metformin Oral Tablet |
- | SCD (Semantic Clinical Drug): Metformin hydrochloride 500 MG Oral Tablet [RxCUI: 860975] |
- | |
- | BN (Brand Name): Glucophage |
- | SBD (Semantic Branded Drug): Glucophage 500 MG Oral Tablet [RxCUI: 105267] |
- +---------------------------------------------------------------------------------------------------+
- RxNorm Concept Entities and Relationships
Ingredient (IN): The active chemical moiety responsible for the therapeutic effect (e.g., Metformin , Lisinopril , Atorvastatin ).
Precise Ingredient (PIN): The specific chemical salt, ester, or hydrate form (e.g., Metformin hydrochloride , Lisinopril dihydrate ).
Dose Form (DF): The physical form of the drug product (e.g., Oral Tablet , Extended Release Oral Capsule , Topical Solution ).
Clinical Drug Form (CDF): Combines the Ingredient and Dose Form without specifying strength (e.g., Metformin Oral Tablet ).
Semantic Clinical Drug (SCD): The primary concept representing an unbranded, generic medication. Formula: SCD = Ingredient(s) + Strength(s) + Dose Form \text{SCD} = \text{Ingredient(s)} + \text{Strength(s)} + \text{Dose Form} SCD = Ingredient(s) + Strength(s) + Dose Form Example: Metformin hydrochloride 500 MG Oral Tablet (RxCUI: 860975 ).
Brand Name (BN): The proprietary commercial trade name assigned by the pharmaceutical manufacturer (e.g., Glucophage , Zestril , Lipitor ).
Semantic Branded Drug (SBD): The branded equivalent of an SCD. Formula: SBD = Brand Name + Strength(s) + Dose Form \text{SBD} = \text{Brand Name} + \text{Strength(s)} + \text{Dose Form} SBD = Brand Name + Strength(s) + Dose Form Example: Glucophage 500 MG Oral Tablet (RxCUI: 105267 ).
The National Drug Code (NDC) Challenge in Analytics
In pharmaceutical claims and billing, medications are identified by the FDA National Drug Code (NDC) . However, relying directly on raw NDCs in health data analytics introduces critical data quality issues: +---------------------------------------------------------------------------------------------------+
- | NDC vs. RxNORM IN HEALTH DATA ANALYTICS |
- +-----------------------------------+---------------------------------------------------------------+
- | NATIONAL DRUG CODE (NDC) | RxNORM (RxCUI) |
- +-----------------------------------+---------------------------------------------------------------+
- | 10- or 11-digit commercial SKU | Normalized semantic clinical concept |
- | Identifies package size & labeler | Identifies active chemical ingredient, strength, & dose form |
- | Ephemeral: recycled & retired | Permanent & immutable semantic concept identifier |
- | Over 200+ NDCs for generic 500mg | Exactly ONE normalized Semantic Clinical Drug (SCD) RxCUI |
- | Metformin depending on bottle size| for all generic 500mg Metformin tablets |
- +-----------------------------------+---------------------------------------------------------------+
NDC Formatting Complexities: FDA NDCs are formatted as 10 digits across three configurations: 4-4-2 , 5-3-2 , or 5-4-1 (representing Labeler Code - Product Code - Package Code ). HIPAA transactions mandate converting these into an 11-digit 5-4-2 format by inserting leading zeros. Failure to standardize NDC formats corrupts SQL joins.
NDC Churn and Obsolescence: Pharmaceutical manufacturers frequently retire, re-package, or reassign NDCs. If an analyst tracks medication adherence (e.g., Proportion of Days Covered [PDC] for statins) using static NDC lists, the metric will artificially plummet as NDCs change. RxNorm maps hundreds of disparate, transient NDCs to a single, stable RxCUI , ensuring seamless longitudinal tracking.
5. UMLS (Unified Medical Language System)
Developed and maintained by the National Library of Medicine (NLM), the Unified Medical Language System (UMLS) integrates and maps over 200 disparate biomedical vocabularies, terminologies, and classification systems (including SNOMED CT, LOINC, RxNorm, ICD-10-CM/PCS, MeSH, and CPT) into a unified relational knowledge graph. ┌───────────────────────────────────────────────────┐
- │ UMLS METATHESAURUS (Concept: C0011849) │
- │ "Diabetes Mellitus" │
- └─────────────────────────┬─────────────────────────┘
- │
- ┌─────────────────────────────────┼─────────────────────────────────┐
- ▼ ▼ ▼
- ┌───────────────────┐ ┌───────────────────┐ ┌───────────────────┐
- │ ICD-10-CM │ │ SNOMED CT │ │ MeSH │
- │ E11.9 │ │ 73211009 │ │ D003920 │
- │ (Type 2 Diabetes) │ │ (Diabetes Mellitus│ │ (Diabetes Mellitus│
- └───────────────────┘ └───────────────────┘ └───────────────────┘
- The Three UMLS Knowledge Sources
UMLS Metathesaurus:
The core multi-vocabulary database. Organizes terms by meaning into Concept Unique Identifiers (CUIs) . Every synonymous term, regardless of source terminology or language, is assigned to the exact same CUI.
Atom Unique Identifier (AUI): Unique identifier for every distinct string occurrence within a specific source vocabulary.
String Unique Identifier (SUI): Identifies identical text strings across vocabularies.
Term Unique Identifier (TUI): Identifies canonical lexical term variations.
UMLS Semantic Network:
Provides a high-level, consistent ontological structure consisting of 127 Semantic Types (broad categories such as Disease or Syndrome , Pharmacologic Substance , Anatomic Structure , Body Location or Region ) and 54 Semantic Relationships (hierarchical links like isa , and associative links like treats , causes , location_of , manifestation_of ).
SPECIALIST Lexicon and Lexical Tools:
Natural Language Processing (NLP) resources designed to manage biomedical terminology variation. Includes lemmatization, tokenization, synonym generation, normalization, and spelling variant tools used by NLP pipelines to extract discrete concepts from unstructured clinical text.
6. Comparative Summary of Clinical Terminologies & Ontologies
Terminology / Standard Governance Organization Domain Scope Primary Structural Identifier Core Analytical Use Case
SNOMED CT SNOMED International & NLM Complete Clinical Medicine (Diseases, Findings, Anatomy, Procedures) Concept ID (SCTID) + FSN Point-of-care EHR clinical documentation, CDS rule engines, polyhierarchical cohort extraction
LOINC Regenstrief Institute Laboratory Tests, Vital Signs, Clinical Documents, Survey Scales 6-Part Name + Code (e.g., 2345-7 ) Lab data exchange (HL7 ORU, FHIR Observation), clinical quality measures, SDOH screening
RxNorm National Library of Medicine (NLM) Normalized Clinical Medications (Generic & Branded) RxCUI (e.g., 860975 for SCD) E-prescribing, medication reconciliation, drug interaction alerts, NDC aggregation for longitudinal research
UMLS Metathesaurus National Library of Medicine (NLM) Multi-Terminology Inter-Mapping (>200 Vocabularies) Concept Unique Identifier (CUI) Cross-vocabulary mapping, biomedical Natural Language Processing (NLP), semantic search