Clinical Data Terminology Standards

Overview

Usage

• Needs to code verbatim adverse event text to MedDRA terms.
• Asks which level of MedDRA to use for frequency tables or signal detection.
• Wants to group diagnoses by ICD-10 codes and must choose 3-character vs 4-character granularity.
• Needs to understand SNOMED CT concept relationships for clinical findings or procedures.
• Must select LOINC codes for lab tests or understand panel composition.
• Is mapping between terminology systems (MedDRA ↔ ICD-10, SNOMED CT → ICD-10).
• Asks about regulatory expectations for AE coding in clinical trials.

1. Identifying the terminology system appropriate for the data type and use case.
2. Clarifying the granularity — which hierarchy level to code at and which to report at.
3. Applying the mapping rules specific to the source and target systems.
4. Flagging common pitfalls — overcoding, undercoding, inappropriate aggregation.
5. Stating regulatory context when relevant (ICH E2B, FDA, EMA expectations).

Response Format

• Lead with the direct recommendation or classification (≤3 sentences)
• Structure as: recommendation → justification (citing specific criteria/thresholds) → caveats
• Use tables for comparisons; bullet points for criteria lists
• Omit background the user already knows — they asked the question
• Target: 200-400 words unless the user requests exhaustive detail The decision trees and frameworks in this skill are for internal reasoning only. Apply them to reach your conclusion, but do not reproduce them in your response. Present only the final recommendation with supporting evidence.

Core Concepts

1. MedDRA: The 5-Level Hierarchy

The coding path

Verbatim AE text (from CRF)
  │
  ▼
LLT (closest match to verbatim)
  │
  ▼
PT (standard reportable term — one LLT maps to exactly one PT)
  │
  ▼
HLT → HLGT → SOC (for grouping and display)

1. Always code to LLT first, then let the hierarchy derive the PT. Never skip to PT directly from verbatim text — the LLT layer captures synonyms and spelling variants.
2. PT is the primary analysis level. Frequency tables in clinical study reports (CSRs) list AEs at the PT level, grouped by SOC.
3. One LLT → exactly one PT. But one PT can belong to multiple SOCs (primary SOC assignment determines the default grouping).
4. Primary SOC assignment follows the ICH convention: etiology SOC takes precedence over manifestation SOC. "Drug-induced hepatitis" goes under "Hepatobiliary disorders" (manifestation) but its primary SOC is determined by MedDRA's primary path rules.
5. Multi-axiality: A PT can appear under multiple SOCs. Always use the primary SOC for frequency tables unless the protocol specifies otherwise.

When to aggregate above PT

Decision tree: MedDRA coding level

Start
│
├── Is this a standard CSR AE table?
│     └── Code at LLT → report at PT within primary SOC.
│
├── Are you looking for a safety signal across related terms?
│     ├── Does an SMQ exist for the topic?
│     │     └── Yes → Use the SMQ (narrow or broad scope as appropriate).
│     └── No SMQ → Aggregate at HLT. Document the grouping rationale.
│
├── Are individual PTs too rare to detect a signal?
│     └── Consider HLT or HLGT grouping, but report both grouped
│         and individual PT counts for transparency.
│
└── Executive safety summary only?
      └── SOC-level counts are acceptable, with PT detail available on request.

2. ICD-10: Code Structure and Grouping

Structure

Chapter (I–XXII)
  └── Block (e.g., E10–E14: Diabetes mellitus)
        └── Category — 3 characters (e.g., E11: Type 2 diabetes)
              └── Subcategory — 4–7 characters (e.g., E11.6: With complications)
                    └── Extension (e.g., E11.65: With hyperglycemia)

Grouping decisions

Decision tree: ICD-10 grouping level

Start
│
├── Are you estimating disease prevalence or incidence?
│     └── 3-character category. Captures all subtypes.
│         Add block-level grouping if comparing across disease families.
│
├── Are you defining a specific clinical outcome?
│     └── Full code (4+ characters). Be explicit about which
│         manifestations qualify. Document inclusion/exclusion codes.
│
├── Are you building a comorbidity score?
│     └── Use the published code lists (Charlson/Elixhauser).
│         Do not invent your own groupings.
│
├── Are you comparing across countries or ICD-10 modifications?
│     └── Block level. National modifications (ICD-10-CM, ICD-10-GM)
│         diverge at the subcategory level.
│
└── Are you mapping from another system (MedDRA, SNOMED CT)?
      └── Map to 3-character first, then refine to subcategory
          only if the source system provides sufficient specificity.

ICD-10-CM vs ICD-10-WHO

• ICD-10-CM (US Clinical Modification): extends codes to 7 characters, adds laterality, encounter type (initial/subsequent/sequela). Used in US claims and EHR.
• ICD-10-WHO: the international base. Shorter codes, no laterality extensions.
• Always specify which modification you are using. A code valid in ICD-10-CM may not exist in ICD-10-WHO.

3. SNOMED CT: Concept Model

Core structure

• Concepts: unique identifiers (SCTIDs) representing clinical meanings.
• Descriptions: human-readable terms (Fully Specified Name, Preferred Term, Synonyms).
• Relationships: typed links between concepts.

Key relationship types

When to use SNOMED CT vs other systems

SNOMED CT hierarchy navigation rules

1. Use the IS-A hierarchy for subsumption queries. To find "all types of diabetes," query descendants of concept 73211009 (Diabetes mellitus).
2. Post-coordination (combining concepts) is powerful but complex. Prefer pre-coordinated concepts when they exist.
3. Reference sets (refsets) define subsets for specific use cases (e.g., a national drug extension). Always check which refset applies to your jurisdiction.
4. SNOMED CT → ICD-10 mapping is maintained by SNOMED International. Use the official map; do not create ad hoc crosswalks.

4. LOINC: Lab Test Coding

The 6-part naming convention

Panels

LOINC selection rules

1. Match all six axes. A glucose test on serum (quantitative, mass concentration) is a different LOINC code than glucose on urine (quantitative, mass concentration) or glucose on serum (ordinal).
2. Prefer codes without a method axis unless the method materially affects interpretation (e.g., calculated vs direct LDL).
3. Use panel codes when the full panel was ordered, and individual member codes for the component results within it.
4. Check the LOINC status. Codes can be ACTIVE, DEPRECATED, or DISCOURAGED. Only use ACTIVE codes.
5. LOINC answers (LA codes) provide standardized answer lists for ordinal observations (e.g., positive/negative/indeterminate).

5. Cross-System Mapping Decisions

MedDRA → ICD-10

1. One MedDRA PT may map to multiple ICD-10 codes (and vice versa). Document the mapping table and version.
2. Never map LLT directly to ICD-10. Always go through PT first — LLTs are synonyms, not distinct concepts.
3. Granularity mismatch is expected. MedDRA PTs are often broader than ICD-10 subcategories. Accept the loss of specificity or require additional clinical context.

SNOMED CT → ICD-10

• Use the SNOMED International maintained map (released with each SNOMED CT edition).
• The map provides target ICD-10 codes with map rules and priorities.
• One SNOMED CT concept may map to multiple ICD-10 codes (combination coding). Follow the map rules to select the correct target(s).
• Do not reverse the map (ICD-10 → SNOMED CT) without a dedicated reverse crosswalk — the forward map is many-to-one and loses information when inverted.

LOINC → local lab codes

• Most institutions maintain a local-to-LOINC mapping table (often called a "LOINC mapping" or "result code crosswalk").
• Validate mappings by checking all six LOINC axes against the local test definition. A common error is mapping a serum test to a urine LOINC code because the component name matches.
• Panels vs components: Map the panel order to the panel LOINC code, and each result line to its individual component LOINC code.

Decision tree: which mapping approach?

Start
│
├── Source is verbatim AE text?
│     └── Code to MedDRA (LLT → PT). Map to ICD-10 only if
│         the downstream system requires it.
│
├── Source is EHR diagnosis (free text or local code)?
│     ├── US system → Map to ICD-10-CM (billing) and SNOMED CT (clinical).
│     └── Non-US → Map to ICD-10-WHO or national modification.
│
├── Source is lab result?
│     └── Map to LOINC. Match all six axes. Use panel codes for orders.
│
├── Need to cross-walk between MedDRA and SNOMED CT?
│     └── Go through ICD-10 as an intermediary (MedDRA → ICD-10 → SNOMED CT)
│         or use UMLS concept mappings. Direct MedDRA ↔ SNOMED CT maps
│         are not officially maintained.
│
└── Building a research dataset combining claims + EHR + trial data?
      └── Define a canonical code system for each data element.
          Map all sources to the canonical. Document every mapping
          decision and version. Accept that some mappings are lossy.

6. Regulatory Context

When NOT to Use This Skill

• Assigning MedDRA codes for regulatory safety reports (needs trained medical coder)
• When coding requires clinical judgment about causality assessment
• ICD-10 coding for billing/reimbursement where legal liability applies

When to Escalate to a Human Expert

• When a term maps to multiple PT candidates with different SOCs
• Cross-walk decisions that affect regulatory submission acceptance
• When local coding conventions override international standards

Common Mistakes

• Wrong: Coding adverse events directly to PT, skipping the LLT layer Right: Always code verbatim text to LLT first, then let the hierarchy derive the PT Why: The LLT layer captures synonyms and spelling variants — skipping it risks inconsistent coding when different verbatim texts describe the same concept
• Wrong: Using secondary SOC for AE frequency tables Right: Always use the primary SOC for frequency tables unless the protocol explicitly specifies otherwise Why: Multi-axial PTs appear under multiple SOCs — mixing primary and secondary inflates apparent AE counts and confuses signal detection
• Wrong: Aggregating at HLT level without documenting justification Right: Only aggregate above PT when individual terms are too sparse for signal detection; always show both grouped and ungrouped PT counts Why: HLT grouping hides individual PT signals that may represent distinct safety concerns
• Wrong: Applying ICD-10-CM codes in a non-US context Right: Specify which ICD-10 modification you are using; use ICD-10-WHO or the appropriate national modification for non-US data Why: ICD-10-CM extensions (laterality, encounter type) do not exist in ICD-10-WHO — codes may be invalid or misinterpreted
• Wrong: Grouping ICD-10 at 3-character level for specific outcome studies Right: Use full codes (4+ characters) when the outcome is a specific manifestation; document inclusion/exclusion codes explicitly Why: E11 includes all Type 2 DM complications — if the outcome is "Type 2 DM with hyperglycemia," only E11.65 is correct
• Wrong: Reversing the SNOMED CT → ICD-10 map to create an ICD-10 → SNOMED CT crosswalk Right: Use a dedicated reverse crosswalk or UMLS CUI-based mappings; never invert the forward map Why: The forward map is many-to-one — reversing it creates ambiguous one-to-many mappings without the map rules to resolve them
• Wrong: Matching LOINC codes by component name alone Right: Verify all six LOINC axes (component, property, timing, system, scale, method) match the local test definition Why: Two LOINC codes can share the same component but differ in system (serum vs urine), property, or scale — partial matching produces wrong assignments
• Wrong: Using deprecated LOINC codes without checking status Right: Always check the LOINC STATUS field and replace deprecated codes with their active successors Why: Deprecated codes break interoperability and may not be recognized by receiving systems
• Wrong: Creating ad hoc terminology crosswalks instead of using official maps Right: Use curated maps from SNOMED International, NLM, or MSSO; document the map version used Why: Ad hoc mappings introduce inconsistencies, are not reproducible, and diverge from regulatory expectations
• Wrong: Ignoring MedDRA version differences when combining data across studies Right: Document the MedDRA version used for each study; reconcile terms that were added, merged, or deprecated across versions Why: PTs are added, merged, and deprecated across versions — a term valid in v25 may be deprecated in v26, causing data loss or miscounting
• Wrong: Confusing LOINC panel codes with component codes when storing results Right: Use the panel code for the order and individual component LOINC codes for each result within the panel Why: Storing a glucose result under the CBC panel code is semantically wrong and breaks downstream aggregation and interoperability
• Wrong: Mapping MedDRA directly to SNOMED CT without an intermediary Right: Use ICD-10 as an intermediary (MedDRA → ICD-10 → SNOMED CT) or use UMLS CUI-based mappings; document the mapping path Why: No official direct MedDRA ↔ SNOMED CT crosswalk exists — direct mapping produces unvalidated, irreproducible results

References

• MedDRA Introductory Guide: https://www.meddra.org/how-to-use/support-documentation
• ICH E2B(R3) guideline: https://database.ich.org/sites/default/files/E2B_R3__Guideline.pdf
• ICD-10-WHO: https://icd.who.int/browse10/2019/en
• ICD-10-CM: https://www.cms.gov/medicare/coding-billing/icd-10-codes
• SNOMED CT Starter Guide: https://confluence.ihtsdotools.org/display/DOCSTART
• SNOMED CT to ICD-10 Map: https://www.snomed.org/snomed-ct/Use-SNOMED-CT/maps
• LOINC Users' Guide: https://loinc.org/kb/users-guide/
• CDISC SDTM Implementation Guide: https://www.cdisc.org/standards/foundational/sdtm
• OMOP CDM Vocabulary: https://ohdsi.github.io/CommonDataModel/