Clinical Data Mappings
Engineer-oriented, illustrative crosswalks across SDTM, ADaM, FHIR, and OMOP — not a substitute for the official IGs.
This is a small, independent cookbook for data engineers who have to move clinical or EHR-shaped data between well-known models: SDTM, ADaM, FHIR, and OMOP CDM v5.4, plus a plain-language note on the MedDRA hierarchy.
Every mapping on this site is illustrative. It is not submission-ready, not validated for a regulatory filing, and not a reproduction of official implementation-guide tables. Examples use synthetic subjects and invented source values only. Licensed dictionaries (MedDRA terms, CDISC Controlled Terminology code lists) are not republished here — official sources are linked instead.
Read the official guides before you ship anything: CDISC, FHIR to CDISC IG, FHIR to OMOP IG, OMOP CDM v5.4, MedDRA.
SDTM DM, AE, and LB — source-shaped pitfalls
Why a perfectly loadable table can still be the wrong grain or the wrong precision for DM, AE, and LB.
FHIR resources to SDTM-shaped rows
An illustrative Patient / AdverseEvent / Observation path into DM-, AE-, and VS-shaped columns, and the gaps official IGs call out.
FHIR to OMOP CDM v5.4 — identity and events
Illustrative PERSON and CONDITION_OCCURRENCE rows from synthetic FHIR-like events, with concept ids left blank on purpose.
MedDRA levels for data engineers
The five-level hierarchy (LLT, PT, HLT, HLGT, SOC) in plain language, without republishing any MedDRA terms or codes.
SDTM to ADaM — ADSL, BDS, and OCCDS
Illustrative derivation flow from SDTM-like domains into analysis-ready structures, without claiming a submission-ready ADaM.
When a value does not fit the parent domain
Supplemental qualifier patterns — why a leftover source column should not be silently dropped or stuffed into a misused variable.
Partial dates and --DTC fields
Why padding a date-only source value with midnight invents precision, shown on a synthetic adverse-event onset.
OMOP vocabulary mapping as a process
How source values become standard concept ids — Athena, source-value retention, and why this site leaves concept ids empty.
Traceability from source to SDTM to ADaM
Identifiers you should not drop, and why a derived analysis column still needs a path back to a source row.
An illustrative end-to-end pipeline
How CDASH-like collection, SDTM-like tabulation, ADaM-like analysis, FHIR exchange, and OMOP analysis relate — and where they do not.